Jun 20, 1997 - virtue of these properties, they can discriminate molecules on the basis .... This is the most widely used important classical technique to identify ... template decomposition products. .... catalytic activity and outstanding resistance to coke formation [65]. .... cations to generate the maximum amount of acid sites.
SYNTHESIS, CHARACTERIZATION AND CATALYTIC PROPERTIES OF AlPO AND VAPO MOLECULAR SIEVES WITH DIFFERENT PORE CHARACTERISTICS
BY N. VENKATATHRI
* [JUNE-1997] *
1
SYNTHESIS, CHARACTERIZATION AND CATALYTIC PROPERTIES OF AlPO AND VAPO MOLECULAR SIEVES WITH DIFFERENT PORE CHARACTERISTICS
A THESIS SUBMITTED TO THE UNIVERSITY OF POONA FOR THE DEGREE OF DOCTOR OF PHILOSOPHY (IN CHEMISTRY)
BY N. VENKATATHRI
CATALYSIS DIVISION NATIONAL CHEMICAL LABORATORY PUNE – 411 008, INDIA
JUNE 1997
2
Dedicated To My Parents
3
CERTIFICATE Certified that the work incorporated in the thesis, ―Synthesis, characterization and catalytic properties of AlPO and VAPO molecular sieves with different pore characteristics‖ submitted by Mr. N. Venkatathri, for the degree of Doctor of philosophy, was carried out by the candidate under my supervision in the Catalysis division, National Chemical Laboratory, Pune 411 008, India. Materials obtained from other sources have duly acknowledged in the thesis.
Dr. S. SIVASANKER Deputy director Catalysis division National Chemical Laboratory Pune 411 008, Maharastra, India
4
ACKNOWLEDGEMENTS I wish to express my sincere gratitude to Dr. S. Sivasanker, Deputy director, National Chemical Laboratory, for his invaluable guidance, through provoking discussions and help rendered throughout the course of this investigation without I would not have completed this thesis successfully. I am deeply indebted to Dr. S. G. Hegde for his valuable discussions, suggestion and professional help rendered during the course of the present investigation. With out his help, it would have been difficult for me to complete my research work successfully. I wish to offer my sincere thanks to Dr. A.V. Ramaswamy, Head, Catalysis division for his stimulating discussions and personal help during the course of my research work without which it would not have been possible for me to present this work in the form of thesis. I am grateful to Drs. K. Vijayamohanan, Veda Ramaswamy, B.S. Rao, Rajiv kumar, A.P. Singh, V.P. Shiralkar, Nalini Jacob, A. Behlekar, G. Sainker, H.S. Soni, R. Vetrivel, A.J. Chandwadkar, Mirajkar, Thamankar, Awate, Agashe, Shinde, Chakraborthy and all other scientific and non-scientific staff in the catalysis division, NCL, Pune for extending their co-operation in all ways to complete my research work successfully. I would be failing in my duty if I don‘t thank my friends especially Vinod, Anil Kumar Sinha, Raghavan, Sujatha, Selvaraj, Gnanaravi, Kumaran, Ramani, Saravanan, T. Selvam, Sasitharan, C.N. Selvam, Tapas sen, Tapan kumar das, Pavan Kumar, A.K. Pandey, N.K. Mal, Ravishanker, Sakthivel, Jayaraman, Mani, Eric Lowenthal, Kannan, Kesavaraj, D. Bhattacharya, Puyam Singh, Bhavana, Sahida, Karuna, Rajaram and other friends for encouraging me throughout the course of work and providing me all the help I needed in time. I also thank Mr. Ramakrishnan for help in making the autoclaves, and Mr. Purushothaman for helping me in many ways during the course of my work. I would take this opportunity to my house owner Misal and family members for their encouragement and support during the final stages of my research work. I thank my brothers, sister, mama, chithappa, nephews, sister-in-laws and all other relatives, who gave full support and encouragement during the course of my research work. I thank the Director, NCL, Pune, Dr. P. Ratnasamy for permitting me to submit this work in the form of a thesis and the Council of Scientific and Industrial Research, New Delhi for providing me a research fellowship. Place: Pune Date: 20-06-1997
Venkatathri. N
5
CONTENTS I.
GENERAL INTRODUCTION 1.1
INTRODUCTION
1
1.2
HISTORY OF ALUMINOPHOSPHATE MOLECULAR
1
SIEVES 1.3
MICROPOROUS ALUMINOPHOSPHATE (AlPO4)
2
MOLECULAR SIEVES 1.4
SYNTHESIS OF ALUMINOPHOSPHATE MOLECULAR
4
SIEVES 1.5
ISOMORPHOUS SUBSTITUTION IN
5
ALUMINOPHOSPHATE MOLECULAR SIEVES 1.5.1
Silicoaluminophosphate (SAPO) molecular sieves
5
1.5.2
Metal aluminophosphate (MeAPO) molecular sieves
6
1.5.3
Metal silicoaluminophosphate (MeAPSO) molecular
7
sieves 1.6
1.7
1.8
PHYSICOCHEMICAL CHARACTERIZATION
8
1.6.1
X-ray diffraction
8
1.6.2
Scanning electron microscopy
8
1.6.3
Infrared spectroscopy
8
1.6.4
Thermal analysis
9
1.6.5
Adsorption studies
10
1.6.6
Electron spin resonance spectroscopy
10
1.6.7
UV-Vis spectroscopy
11
1.6.8
NMR spectroscopy
11
1.6.9
Cyclic voltammetry
11
APPLICATIONS
12
1.7.1
Catalytic dewaxing
12
1.7.2
p-xylene production
13
1.7.3
Olefin Isomerization
13
1.7.4
Conversion of methanol to light olefins
13
1.7.5
Alkylation of aromatics with methanol
13
1.7.6
Octane enhancing catalyst additives
13
STRUCTURE OF SOME ALUMINOPHOSPHATE
15
MOLECULAR SIEVES WITH DIFFERENT PORE CHARACTERISTICS
1.9
1.8.1
AlPO4-16
15
1.8.2
AlPO4-22
16
1.8.3
SAPO-35
17
1.8.4
AlPO4-31
18
1.8.5
AlPO4-41
19
1.8.6
AlPO4-5
20
VANADIUM ALUMINOPHOSPHATGE (VAPO)
21
MOLECULAR SIEVES
II.
2.0
SCOPE OF THE THESIS
23
2.1
REFERENCES
26
EXPERIMENTAL 2.1
INTRODUCTION
32
2.2
SYNTHESIS
32
2.3
PRETREATMENT PROCEDURE
32
2.4
METHODS USED FOR PHYSICOCHEMICAL
36
CHARACTERIZATION 2.4.1
Chemical analysis
36
6
2.4.1A
Estimation of Aluminum as 8-hydroxy
36
quinolate [Al(C9H6OH)3] 2.4.1B
Estimation of phosphate as magnesium
36
phosphate (Mg2P2O7)
III.
IV.
2.4.1C
Estimation of silicon
37
2.4.1D
Estimation of carbon and nitrogen
37
2.4.1E
Estimation of V4+and V5+ ions.
37
2.4.1F
Ion exchange
38
2.4.2
X-ray diffraction
38
2.4.3
Scanning electron microscopy
39
2.4.4
FT-IR spectroscopy
39
2.4.5
ESR spectroscopy
41
2.4.6
UV-Vis spectroscopy
41
2.4.7
Thermal analysis
41
2.4.8
NMR spectroscopy
42
2.4.9
Cyclic voltammetry
42
2.4.10
Catalytic properties
45
2.4.10A
n-hexane oxifunctionalization
45
2.4.10B
Benzaldehyde acetalization
45
2.4.10C
Toluene and Cyclohexene oxidation
45
2.4.10D
Electrocatalysis
46
2.4.10E
Analysis
46
SYNTHESIS 3.1
INTRODUCTION
47
3.2
SYNTHESIS OF AlPO4 AND VAPO MOLECULAR SIEVES
47
3.2.1
Variation in template concentration
47
3.2.2
Influence of Al/P ratio
50
3.2.3
Influence of water content
50
3.2.4
Influence of aluminium source
50
3.2.5
Influence of silica and vanadium oxide addition
52
3.2.6
Influence of crystallization medium
55
3.2.7
Influence of pH
55
3.3
CRYSTALLIZATION KINETICS OF AlPO4-5
58
3.4
REFERENCES
68
PHYSICOCHEMICAL CHARACTERIZATION 4.1
INTRODUCTION
69
4.2
ELEMENTAL ANALYSIS
69
4.2.1
Aluminophosphate molecular sieves
71
4.2.2
Vanadium aluminophosphate molecular sieves
71
4.2.2A
VAPO-5 and VAPSO-5
71
4.2.2B
VAPO-31 and VAPO-41
76
4.3
X-RAY DIFFRACTION
76
4.4
SCANNING ELECTRON MICROSCOPY
77
4.5
THERMAL ANALYSIS
84
4.5.1
Aluminophosphate molecular sieves
84
4.5.1A
AlPO4-L
84
4.5.1B
AlPO4-5, AlPO4-16 and AlPO4-22
84
4.5.1C
SAPO-35
89
4.5.2
4.6
Vanadium aluminophosphate molecular sieves
92
4.5.2A
VAPO-5 and VAPSO-5
92
4.5.2B
VAPO-31 and VAPO-41
92
FOURIER TRANSFORM INFRARED SPECTROSCOPY
7
95
4.7
Framework region vibrations
98
4.6.2
Hydroxyl region vibrations
98
NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY 4.7.1
13
4.7.2
27
C MAS NMR
104 104
Al and 29Si MASNMR
104
4.8
ELECTRON SPIN RESONANCE SPECTROSCOPY
109
4.9
UV-VIS SPSECTROSCOPY
112
4.10
CYCLIC VOLTAMMETRY
112
4.10.1
VAPO-5, VS-1, VS-2 AND V-Al-Beta
116
4.10.2
Other vanadium containing molecular sieves
133
4.10.3
Active site
133
4.11 V.
4.6.1
REFERENCES
138
CATALYTIC PROPERTIES 5.1
INTRODUCTION
141
5.2
TOLUENE OXIDATION
141
5.2.1
141
Oxidation using tertiary butyl hydro peroxide as oxidant
5.2.2
VI.
Electrocatalysis
144
5.3
CYCLOHEXENE EPOXIDATION
144
5.4
OXIFUNCTIONALIZATION OF N-HEXANE
144
5.5
BENZALDEHYDE ACETALIZATION
147
5.6
MECHANISM OF TOLUENE OXIDATION
149
5.7
REFERENCES
153 154
CONCLUSIONS
8
CHAPTER I ________________________________________________________________________ GENERAL INTRODUCTION ________________________________________________________________________
9
1.1: INTRODUCTION Crystalline microporous aluminosilicates (zeolites) and aluminophosphates (AlPO4s) are classified as molecular sieves. While aluminosilicates have been known for about a century [1], aluminophosphates were discovered only about two decades ago [2]. At present more than twenty aluminophosphates are known.
Some are structurally
related to zeolites and some are novel. The family of aluminophosphates has been extended by incorporation of silicon or metal ions or MeAPSO) [3-7].
both (SAPO, MeAPO and
Aluminophosphates are made-up of a rigid three dimensional
framework structure enclosing cavities and channels. AlPO4s are interesting materials because their pores are uniform in size and in the size range of small molecules. By virtue of these properties, they can discriminate molecules on the basis of their size and this ability qualified them as molecular sieves.
1.2: HISTORY OF ALUMINOPHOSPHATES Silicon and aluminium are not unique in their ability to form tetrahedrally coordinated oxide networks. The element phosphorous to the right of silicon in the periodic table frequently assumes tetrahedral co-ordination with oxygen. Aluminophosphates posses many structural similarities to silicon with phosphorous in the +5 oxidation state: i) AlPO4 is isoelectronic with Si2O4, ii) The average ionic radii of Al3+ (0.39 nm) and P5+ (0.17 nm) is 0.28 nm very close to the ionic radius of Si4+ (0.26 nm) [9] and iii) AlPO4 and SiO2 form isomorphic dense phases with Al3+ alternating with P5+ in a tetrahedral oxide framework. There are dense phases of AlPO4 corresponding to seven structural forms of SiO2, alfa, and beta quartz, alfa, beta and gamma tridymite and Alfa and beta crystobolite [10]. These structural similarities between AlPO4 and SiO2 served as a stimulus for examining aluminophosphates as potential source of microporous framework structures.
Two hydrated meta stable forms of AlPO4.2H2O are varscite and meta
varscite composed of alternating aluminium and phosphorous exhibiting octahedral and tetrahedral co-ordinations respectively [11]. 1.3: MICROPOROUS ALUMINOPHOSPHATE (AlPO4) MOLECULAR SIEVES A new generation of molecular sieves, aluminophosphates (AlPO4s) were reported by Wilson et al. in 1982, having framework compositions comprising of [AlO2]and [PO2]+ tetrahedral [2-4]. Aluminophosphate molecular sieves are the first reported novel class of crystalline microporous oxide framework structures synthesized without silica [2,8]. These molecular sieves are similar to zeolites in some properties and it has been claimed that these may be used as adsorbents, catalysts and catalyst supports. The new AlPO4 family currently includes about twenty two, three dimensional framework structures (Table 1:1) of which at least sixteen are microporous and six are two dimensional layer type materials. Most of these three dimensional structures are novel, e.g. AlPO4-5. 31, 41 and 22. Some of the aluminophosphates are structurally similar to zeolites, e.g. AlPO4-17 (erionite), AlPO4-20 (sodalite), AlPO4-24 (analcime), AlPO4-34 (chabazite) and SAPO-35 (levynite). The aluminophosphate molecular sieves exhibit intracrystalline pore volumes for H2O from 0.04 to 0.35 cm3/g and adsorption pore sizes from 0.3 to 1.21 nm covering the entire range of pore volumes and pore sizes known in zeolites [3] and silica molecular sieves. The pure silica molecular sieves, silicalites and neutral SiO2 framework, having no extra framework cations are hydrophobic. In contrast, the neutral aluminophosphate frameworks with no extra framework cations are moderately hydrophilic. The difference may probably be due to the difference in electronegativity between Al(1.5) and P(2.1). They exhibit less affinity for H2O than the hydrophilic zeolites such as type A and type X., but substantially more than the hydrophobic silicalite. Their framework composition with an Al/P ratio of 1 has a wide structural diversity.
10
Aluminophosphates are classified on the basis of their pore diameter, into five major groups, viz. small, medium, large and extra large pore molecular sieves (Table 1: 1) according to the largest membered ring (MR) present in the structure of each AlPO4. No systematic nomenclature has been developed for molecular sieve materials. The discoverer of the novel synthetic molecular sieve, identified based on characteristic X-ray powder diffraction pattern and chemical composition, generally assigns trivial symbols.
The International Zeolite Association-structure commission (IZA) and
International Union of Pure and Applied Chemistry (IUPAC) have assigned structural codes to known synthetic and natural aluminophosphates for known framework topologies irrespective of composition, distribution of T-atoms and unit cell symmetry, e.g. AFI for AlPO4-5, AEL for AlPO4-11 and ERI for AlPO4-17. 1.4: SYNTHESIS OF ALUMINOPHOSPHATES Aluminophosphates are synthesized hydro thermally in the temperature range 373-531 K from a reaction mixture containing sources of aluminium, phosphorous and an organic amine or a quaternary ammonium salt [13] which gets entrapped or clathrated within the crystalline products. The gel molar composition used generally is Al2O3: P2O5: 1-3 R: 40-100 H2O (where R= template) The synthesis procedure involves the following steps: -aluminophosphate gelation, -aging -formation of active gel, -crystallization, -washing, -drying and Physicochemical characterization. The elemental composition of the synthesized molecular sieves is XR: Al2O3: P2O5: y H2O, the quantities x and y representing the amount of organic compound and water needed to fill the microporous voids within the neutral aluminophosphate framework. Synthesis without R gives rise to dense aluminophosphate structures of known hydrates (AlPO4.nH2O). Since the aluminophosphate framework is electrically neutral, the template is not needed as a charge balancing agent; therefore its incorporation into the structure is a function of its electronic nature, size and shape relative to the channel volume to be filled. The entrapped organic species are moved by thermal decomposition. Moreover, these novel structures exhibit thermal stability. Most of these remain crystalline after calcination around 673-873 K, which is necessary for the removal of the organic template and the intracrystalline void volume free for adsorption and catalysis. A single template can give rise to multiple AlPO4 structures and a variety of structures can be synthesized from a single template. AlPO4-5 is synthesized using a large variety of organic templates [2]. One the contrary, AlPO4-16 has been synthesized with only one template, namely quinuclidine [2]. 1.5
:
ISOMORPHOUS
SUBSTITUTION
IN
ALUMINOPHOSPHATE
MOLECULAR SIEVES 1.5.1: Silicoaluminophosphate (SAPO) molecular sieves Silicoaluminophosphates exhibit structural diversity with some 13 three dimensional microporous framework structures. These include SAPO-35 (Levynite), SAPO-35 (faujasite), SAPO-42 (zeolite A), etc. Novel structure types topologically related to the aluminophosphates are SAPO-5, SAPO-11, SAPO-16, SAPO-31 and SAPO-20. SAPO molecular sieves have tetrahedral oxide frameworks containing silicon, aluminium and phosphorous. Mechanistically, their composition can be visualized in terms of silicon substitution into hypothetical aluminophosphate frameworks, the substitution can be [14] i) silicon for aluminum or ii) silicon for phosphorous or iii) simultaneous substitution of two silicon atoms for aluminium and phosphorous. The net 11
framework charge per framework silicon atom resulting from each mode of substitution would be +1 for (i), -1 for (ii) and ) for (iii). Thus some of these materials will have anionic frameworks with a net negative charge coupled with exchangeable cations and Bronsted acid sites.
SAPO molecular sieves exhibit a range of moderate to high
hydrophilic surface properties because of the variable presence of cations and hydroxyl groups and the local electronegativity differences between Si, Al and P. The materials can be used as adsorbents, catalysts and ion exchangers. Their adsorption properties, pore sizes, thermal and hydrothermal stabilities resemble those of AlPO4 molecular sieves.
The acidic properties vary from mild to strong depending on the silicon
concentration and structure type. 1.5.2: Metal aluminophosphate (MeAPO) molecular sieves As aluminophosphate molecular sieves are neutral, Pyke et al [14] attempted to induce useful catalytic activity by chemical modification, adsorption or entrapment of catalytic species with in the pore system.
However, direct lattice substitution of
equivalent or altervalent cations offers a mechanism for introducing ion exchange ability and acidic or basic centers.
Hence di, tri and tetravalent cations which can adopt
tetrahedral co-ordination could be incorporated during hydrothermal synthesis. Such attempts have been [15] made to introduce catalytically active centers by incorporating Si4+, Ti4+, Sn4+, Fe3+, B3+, Ni2+, Co2+, Mg2+, Sr2+, La3+ and V5+ during the hydrothermal synthesis. The novel modified microporous crystalline aluminophosphates (MeAPO-n) have pore diameters ranging from 3 to 10 Ao. The substitution causes the formation of acid sites after calcination. Their framework composition contains metal, aluminium and phosphorous. They exhibit a wide range of composition within the general formula, 00.3R (MexAlyPz)O2. The value of x (mole fraction of Me) typically varies from 0.01 to 0.25. The introduction of divalent ions into a lattice made up of tetrahedral units of AlO2and PO2+ creates acid sites and ion exchange capacity. In this mode of substitution in the MeAPO the metal appears to substitute Al rather than P. Divalent or trivalent metal for trivalent aluminium results in a net negative charge or a neutral framework charge respectively.
Like SAPO-n the negatively charged MeAPO framework possess ion
exchange properties and potential Bronsted acid sites. MeAPO-n also shows structural diversity and compositional variation as regards isomorphous substitution. Thirteen structure type crystallized in the MeAPO family include framework topologies related to the zeolites and novel structures e.g. AlPO4-5 and AlPO4-11. Adsorption, pore size, pore volume and hydrophilic surface selectivity of the MeAPOs are similar to those of SAPOs and AlPO4s.
The observed catalytic
properties are dependent on both metal and structure type. The thermal and hydrothermal stability of the MeAPO-n material is some what less than that of AlPO4 and SAPO materials. 1.5.3: Metal silicoaluminophosphate (MeAPSO) molecular sieves MeAPSO [16-18] structure types include framework topologies observed in binary (AlPO40 and ternary (SAPO and MeAPO) compositional systems of MeO2, AlO2, PO2 and SiO2 tetrahedral units. The empirical chemical composition on anhydrous basis for these molecular sieves is expressed as mR (MewAlxPySiz)O2, where w,x,y and z represent the molar fractions of Me, Al, P and Si respectively, present as tetrahedral oxides and m may vary from 0 to 0.03. Some molecular sieves with ZnO2, AlO2, PO2 and SiO2 and other molecular sieves with TiO2, AlO2, PO2 and SiO2 oxide framework structures have been reported by Lok et al [19]. When additional elements such as Li, Be, B, Ga, Ge, As and Ti are incorporated into the framework structures, many large, intermediate and small pore molecular sieves are formed. These molecular sieves are designated as ElAPO-n [20-22] and ElAPSO-n [16-19]. Quinary and senary framework compositions have been synthesized containing aluminium, phosphorous and silicon with additional combination of divalent metals. 12
1.6: PHYSICOCHEMICAL CHARACTERIZTION 1.6.1: X-ray diffraction (XRD) This is the most widely used important classical technique to identify crystalline solids and to study the kinetics and mechanism of zeolite crystallization [23]. Variations in the lattice parameters and framework symmetry [24,25], collapse of crystal structure and presence of alien phases are also detected by X-ray diffraction measurements. 1.6.2: Scanning electron microscopy (SEM) Scanning electron microscopy (SEM) is a technique used to measure the particle size distribution and morphology of the crystals.
The amorphous materials
present can be detected from scanning electron microscopy. 1.6.3: Infrared spectroscopy (IR) Infrared spectroscopy is an important tenchnique for the investigation of zeolite framework vibrations and is complementary to X-ray structure analysis. The fundamental vibrations of ALPO4/SiO4 terahedra are in the mid infra red region (200-1300cm-1) and these have been in conjunction with XRD to identify the zeolite structure [26]. It is found that the main Si-O, Al-O vibrational band occurs at about 1100cm-1 and is related to the Si/Al ratio in the zeolite framework [27]. Crystallization of the zeolites during the synthesis has been studied by monitoring the changes in the framework IR vibrational frequencies and comparing these with those of the reactants [28]. It is also possible to study the progressive incorporation of the organic cation into the zeolite lattice by measuring the intensity of the bands characteristic of the organic cation.
The IR
technique has been extensively used to distinguish the different types of structural hydroxyl groups from other types of hydroxyl groups, for instance an OH- group attached to the cation in cation exchanged zeolite formed by the hydrolysis of water. The OHgroups are characterized by adsorption bands in the region 3500-3750cm-1 . IR [29] is also used for detection and estimation of acid sites by adsorption of bases and for the study of the transformation of Bronsted of Lewis acid sites occurring at high temperatures or under specified conditions. FT-IR (Fourier transform infrared) spectroscopy has been used to show [30] that most of the templating species are removed on heating the material leaving behind the degradation products in the inorganic matrix. Hedge et al. [31] have investigated the acidic sites of ALPO4-5 and SAPO – 5 by using diffuse reflectance IR spectroscopy. They found that SAPO-5 contains stronger Bronsted acid sites than ALPO4-5. The structural OH groups differ in their acidic properties and this has been investigated in the case of ALPO4-5 and SAPO-5 by means of IR spectroscopy with adsorption of ammonia or pyridine as probe molecules [31]. 1.6.4: Thermal analysis This technique has been used [32,33] to obtain information on the synthesis, mechanism as well as the thermal behavior of synthesized zeolites and identification of template decomposition products. Various physiochemical changes occurring during the thermal treatment are reflected in the DTA/TG/DTG curves. Dehydration of adsorbed water, decomposition of occluded organic cations and dehydroxylation at higher temperatures to produce Lewis acid sites are reflected in these thermo analytical curves. The location of water molecules in hydrated zeolites has been identified by analyzing the multiple endothermic (DTA) peaks due to dehydration [34].
TG-MS provides
information on the nature of the template decomposition products. Thermal analysis of molecular sieves finds extensive applications in studying the kinetics of water desorption [35], oxidative thermal decomposition of organic templates occluded in the voids and thermal stability of the zeolite structure.
13
1.6.5: Adsorption studies The molecular sieving properties of zeolites are usually determined by their pore dimensions and characterized by their sorption capacities for various sorbate molecules. The molecular exclusion properties of these zeolites have been used to estimate the size and shape of the pores in zeolites and to compare them with those determined by structural analysis [36]. Sorption studies often help in determining the specific interactions between the sorbate and the sorbent. Isosteric heats of adsorption reflects the nature and strength of the sorbate-sorbent and sorbate – sorbate interactions [37,38]. Void volumes in zeolites can be measured by adsorption of molecules which have free access into the voids [39]. Derouane [40] and Gabelica et al. [41] have studied the adsorption of n-hexane to investigate the influence of zeolite modification on sorption properties. The void pore volume of zeolite solids is often determined by low temperature (77K) nitrogen sorption. Analysis of such sorption isotherms have been found to be useful in determining the micropore volume and pore size distribution of composite materials containing molecular sieves [42-43]. 1.6.6: Electron spin resonance spectroscopy (ESR) ESR spectroscopy is a sensitive tool to study the environmental symmetry of paramagnetic metal ions.
ESR spectroscopy is used to study the environment of
isomorphously substituted and exchanged cations and metal oxides within the pores of the zeolite or on the external surface of the crystallites [44,45]. Various paramagnetic metal ions such as V4+, Fe3+ and Cr3+ have been studied extensively using ESR spectroscopy [46-48]. The ESR technique [47] has been used to characterize solid-solid reactions between the zeolite and the metal oxide of elements such as Cu, Cr, Fe, V and Mo. The oxidation state, environment and redox property of V in vanadium silicate molecular sieves with MFI and MEL structures have been studied [48-50]. 1.6.7: UV-Vis spectroscopy UV-Visible spectroscopy gives information on the co-ordination of metal complexes. It has been used to study the co-ordination and nature of the vanadium species in vanadium containing molecular sieves [51]. 1.6.8: Nuclear magnetic resonance spectroscopy (NMR) High resolution solid state MASNMR spectroscopy has emerged as a powerful technique for the investigation of zeolite lattice structures and location of various isomorphously substituted metals ions [52]. NMR is very sensitive to the local ordering and geometry. Till to date about twenty NMR active nuclei in zeolites [53] have been studied by the above technique. It is also used to investigate Si and Al ordering [54] in the zeolite framework and to identify the crystallographically equivalent [55] and nonequivalent [56] Si and Al ions at various sites. It has also been used to calculate framework Si/Al ratios [57] and the co-ordination number of Si [58] and Al [59]. NMR spectral correlations have been established with Si-O-T [60] bond angles and Si-O- bond lengths [61].
13
C – NMR spectroscopy is useful in identifying template carbon molecule
interaction with the framework of the zeolite [62]. 1.6.9: Cyclic voltammetry Electrochemical techniques are highly useful in characterizing the local environment (i.e. co-ordination number as well as the oxidation state) of the metal ions particularly if they are located at the surface of the network structure. For example recent studies [63,64] On the redox characteristics of the Ti4+ ions in TS-1 and TS-2 by cyclic voltammetry indicate that only the Ti4+ ions present in the framework positions of molecular sieves readily undergo electron transfer processes, unlike the extra framework ions. All the Ti4+ ions have similar redox properties leading to the observation of a single redox peak in cyclic voltammetric experiments.
14
1.7: APPLICATIONS The aluminophosphate based molecular sieves are active and selective catalysts for a variety of important hydrocarbon conversions.
As acid catalysts the SAPO‘s
promote olefin summarization and oligomerization, but they are less effective for skeletal isomerization.
As acid components of bifunctional catalysts, they are selective for
paraffin and cycloparaffin isomerization with low cracking activity. In general, SAPOs are considerably less active than LZ-105 and other zeolites for olefin, paraffin and aromatic conversion when compared at the same temperatures but they are more selective for olefin and paraffin isomerization when evaluated at comparable conversions. The introduction of transition metals into the framework positions enhances the activity and selectivity for olefin oligomerization relative to plain SAPOs. The MeAPOs and MeAPSOs are also more selective for C8 aromatic rearrangement reactions, an effect that cannot be attributed solely to improved shape selectivity. The enhanced selectivity may be due to uniquely located acid sites in the transition metal containing molecular sieve combined with unexpected spatial effects. It may also be due to a ligand or electronic effect of the transition metal that affects the transition states in aromatic isomerization and disproportionation reactions. Some of the most promising applications are enumerated below. A few of these applications are expected to become commercial in the near future. 1.7.1: Catalytic dewaxing Because of the characteristically high yield in reactions with paraffins and olefins, catalytic dewaxing is a typical application in which aluminophosphate compositions show outstanding selectivity in the conversion of heavy petroleum products [65]. Whereas, in conventional catalytic dewaxing, zeolites typically crack the paraffins to lighter olefinic materials, aluminophosphates isomerizes and long chain paraffins to nonwaxy hydrocarbons of similar molecular weight. The result is an incremental yield advantage and high product quality [65]. 1.7.2: p-xylene production Certain aluminophosphate catalysts show high selectivity, particularly high xylene yields in the conversion of ethyl benzene containing C8 aromatic mixtures, high catalytic activity and outstanding resistance to coke formation [65]. These characteristics provide opportunity for commercial application as catalysts in the production of p-xylene from C8 aromatics. 1.7.3: Olefin isomerization The low coking tendency of medium pore aluminophosphate molecular sieves has resulted in the first viable approach to the skeletal isomerization of olefins, such as the isomerization of n-butene to isobutene [66]. 1.7.4: Conversion of methanol to light olefins Although ZSM-5 zeolites have been proposed for this application, recent work has shown that non-zeolitic molecular sieves like SAPO-34 can produce ethylene and propylene at high selectivity and still remain active (coke free) over long periods of time. The selectivity to ethylene reportedly can be enhanced by the use of MeAPSO compositions, where the metal consists of either Ni or Co [67]. 1.7.5: Alkylation of aromatics with methanol Methyl substituted aromatic compounds can be produced by the alkylation of aromatics with methanol. In particular, p-xylene can be selectively produced by the shape selective alkylation of toluene with methanol over a medium pore aluminophosphate catalyst [68]. 1.7.6: Octane enhancing catalyst additives Medium pore aluminophosphate molecular sieves show high selectivity toward isomerization over cracking when processing paraffins and olefins. This characteristic
15
feature enables the use of aluminophosphates as catalyst additives. These applications are similar to the use of ZSM-5 in FCC applications [69]. 1.8: STRUCTURES OF SOME ALUMINOPHOSPHATE MOLECULAR SIEVES WITH DIFFERENT PORE CHARACTERISTICS AlPO4-16, AlPO4-22, SAPO-35, AlPO4-31, AlPO4-41 and AlPO4—5 are aluminophosphate molecular sieves with small , medium and large pore openings. The typical structural properties of these aluminophosphates are described below. 1.8.1.: AlPO4-16 The framework topology of AlPO4-16 can be described as an idealized zunyite framework. It is built from double four rings. The largest ore opening is a six member ring [70]. Chemical composition Al20P20O80 Symmetry
Cubic
Unit cell constants
a = 13.4 Ao
Space group
F23
Pore structure
[001]8MR; 3Ao
Fig. 1.1: Structure of AlPO4-16: framework viewed along [001].
16
1.8.2. : AlPO4-22 The structure of AlPO4-22 contains two new polyhedral units identified as RPA and AWW units. The AWW cages have two topologically distinct four rings and four six rings. Each AWW cage encapsulates a phosphate tetrahedron. The presence of the occluded phosphate results in P/Al ratio greater than unity. The RPA cage is composed of eight four rings, eight six rings and two eight rings and contains the quinuclidine molecule [71]. Chemical composition
[Al24P24O96] (HPO4)2
Symmetry
Tetragonal
Unit cell constants
a = 13.6 and c = 15.5Ao
Space group
P4/ncc
Pore Structure
[001]8MR; 3.9 Ao
Fig. 1.2: Structure of AlPO4-22: a)framework viewed along [001]; (b) 8 membered ring viewed along [001].
17
1.8.3. SAPO-35 The framework of SAPO-35 is isostructural to levyne. The framework of levyne can be described as alternate layers of hexagonal prisms and single six member rings, alternating such that every seventh layer is superimposable upon the first [72]. Chemical composition Al26.2P23.7Si4.2O108 Symmetry
Trigonal
Unit cell constants : a = 13.3; c = 23.0 Ao Space group
: R3m
Pore Structure : [001]8MR; 3.6 x 4.8 Ao
Fig. 1.3: Structure of SAPO-35: (a) framework viewed along [001]; (b) 8 membered ring viewed along [001].
18
1.8.4. AlPO4-31 The AlPO4 -31 structure consist of non-planar layers of 42, 62, 12-rings linked with staggered four rings. The pore size predicted from adsorption pore gauging is 6.5Ao. Entry in to the structure is through 12 membered rings [73]. Chemical composition : Al18P18O72 Symmetry : Hexagonal Unit cell constants : a=b=20.3; c = 5.0 Ao Space group : R3 Pore Structure : [001] 12MR; 5.4 Ao
Fig. 1.4: Structure of AlPO4-31: (a) framework viewed along [001]; (b) 12 membered ring viewed along [001].
19
1.8.5. AlPO4-41 The framework topology of AlPO4-41 consists of 4.62.5.10 layers connected by U.D.U.D. chains. AlPO4-41 exhibits properties of medium pore molecular sieves [74]. Chemical composition: Al20P20O80 Symmetry: Orthorhombic Unit cell constants : a = 9.73; b = 25.83 Ao Space group = Cmc21 Pore structure : [001] 10MR; 5.70 x 4.3 Ao
Fig. 1.5: Structure of AlPO4-41: (a) framework viewed along [001]; (b) 10 membered ring viewed along [001].
20
1.8.6. AlPO4-5 AlPO4-5 contains alternating Al and P atoms forming four and six membered rings. The pore system consists of non connecting parallel channels of 12 member rings. In the as-synthesized form, this molecular sieve contains tetrapropylammonium hydroxide species when prepared using the TPAOH as the organic additive. The tripod orientation of the TPAOH in the channel system of this aluminophosphate differs significantly from that observed in silicalite, where the propyl groups radiate from the channel intersection into the channel openings [75,76]. Chemical composition : Al12P12O48 Symmetry : Hexagonal Unitcell constants: a = 13.726; c = 8.484 Ao Space group : P6cc Pore Structure: [001]12MR; 7.3Ao
Fig. 1.6: Structure of AlPO4-5: (a) framework viewed along [001]; (b) 12 membered ring viewed along [001].
21
1.9: VANADIUM ALUMINOPHOSPHATE (VAPO) MOLECULAR SIEVES
Vanadium oxide catalysts are industrially important in a number of catalytic processes such as the selective oxidation of hydrocarbons, production of SO3, ammoxidation of hydrocarbons and reduction of nitric oxide. Normally silica, alumina, and titania are used as supports [77,78]. The activity and selectivity of the vanadium oxide catalysts are mainly attributed to the changes in the oxidation state during the reaction. The development of vanadium containing materials with novel structures and well defined states of vanadium is an important area of research in oxidation catalysts.
Vanadium substitution in various silicate molecular sieves, namely, VS-1 [79,80], VS-2 [81], V-ZSM-48 [82], V-MCM-41 [83], and V-ZSM-22 [84] were recently reported in detail. Vanadium substitution for phosphorous in oxides has been reported [86,87]. For example, the replacement of (VO4)-3 is known to occur in the mineral schoderite (2Al2O3: V2O5: P2O5: 16H2O) [85].
The earliest incorporation of vanadium in
aluminophosphates (in AlPO4-5 and AlPO4-11) was patented by Flanigen et al. [86]. Pyke et al.10 reported the same in the case of AlPO4-21. Later Miyamato et al. [88] reported that VAPO-5 framework was more effective than pentasil silicalite framework to incorporate vanadium ions and to provide vanadium sites for the selective ammoxidation of propane.
Montes et al. [89] have reported that vanadium replaces
phosphorous in AlPO4-5 structure and is located in defect sites. They also found that assynthesized VAPO-5 molecular sieves contain vanadyl species in square pyramidal coordination.
The intensity of the ESR lines almost disappeared on calcination and
reappeared on thermal reduction. Jung et al. [90] suggested that vanadium replaces phosphorous in AlPO4-5 framework. They estimated the V4+ and V5+ ions present in the as-synthesized samples. The V5+ content in the as-synthesized samples increased with vanadium concentration.
From a UV-Vis. Analysis, they found the presence of
tetrahedral vanadium in the as-synthesized and calcined samples. Kornatowski et al. [91] synthesized large
660m dia VAPO-5 crystals and
studied the incorporation of vanadium into the AlPO4 framework. Haanepan et al. [92] found higher selectivity for epoxidation of some allyl alcohols (using tertiary butyl hydro peroxide oxidant) on VAPO-5 and VAPO-11 molecular sieves. Rigutto et al. [93] have suggested that vanadium replaces aluminium and not phosphorous in AlPO4-5 framework. Based on NMR and ESR analysis they reported the location of vanadium in square pyramidal symmetry in as-synthesized as well as calcined samples. Also they found a good activity and selectivity towards epoxidation of simple alkenes and oxygenation of alkyl aromatics using tertiary butyl hydroperoxide oxidant. Blasco et al. [94] reported that VAPO-5 samples are active and selective catalysts for the oxidative dehydrogenation of propane.
Both the activity and selectivity obtained over these
materials were higher than over AlPO4-5 and V2O5 supported AlPO4-5. The activity increased linearly with the vanadium content up to about 2 wt%. Isolated tetrahedral V5+ species were proposed to be the active and selective sites for the oxidative dehydrogenation of propane. Weckhnysen et al. [95] also carried out studies on the synthesis and characterization of many VAPO‘s. 2.0: SCOPE OF THE THESIS Silicoaluminophospahate molecular sieves are developing into a class of industrially important catalyst (96). They are most useful only when the P 5+ ions in the framework are replaced by Si4+ cations to generate the maximum amount of acid sites. Some of the problem faced in the incorporation of Si in AlPO4 lattice is insufficient and non specific Si4+ incorporation and poor crystallinity of the SAPOs [4, 97,98,104,105]. Numerous efforts have been made in the past to overcome these problems [99-101], notably synthesis in nonaqueous medium like ethylene glycol, n-hexanol etc. and using novel structure directing templates [102,103].
In the process, a few novel AlPO4
molecular sieves were also obtained [110]. The role of organic cations and amines 22
present in aluminosilicates and aluminophosphates can be as pore filing or structure directing agents.
Often, cations with very flexible conformation possibilities like
tetraethyl ammonium hydroxide (TEAOH) are used. The TEAOH molecule is so flexible that it can produce different structure types like ZSM-5, ZSM-12, beta, SAPO-34 and VPI-8. The compound, hexamethyleneimine (HEM) has been used in the synthesis of PSH-3 or MCM-22 and naonacil [106-109], but no example of AlPO4 synthesis has been reported. HEM is a reasonably flexible cyclic compound (I) and it is interesting to use it in the synthesis of AlPO4s and related materials. In the present work the usefulness of HEM in hydrothermal synthesis of AlPO4 molecular sieves is established.
By varying the synthesis medium, reactants and
parameters many AlPO4 molecular sieves, viz., AlPO4-5, AlPO4-22, AlPO4-16, SAPO35, VAPO-5, VAPSO-5, VAPO-31 and VAPO-41 have been obtained. These materials have been characterized by conventional techniques like XRD, SEM, FT-IR, MASNMR, UV-Vis, ESR, TG/MS, TG/DTA and adsorption techniques. Cyclic voltammetry was employed for the first time in characterizing V-molecular sieves. The acidic and redox properties were characterized by catalytic activity in acetalization of benzaldehyde and toluene
oxidation,
electrocatalysis,
cyclohexene
epoxidation
and
n-hexane
oxyfunctionalization reactions. This work is presented in four more chapters The second chapter provides a description of the various experimental techniques adopted in the hydrothermal synthesis of the microporous materials under investigation and the various characterization techniques used in the present study. The methods adopted for the evaluation of the catalytic properties of the samples are also described. The third chapter described the use of hexamethyleneimine template in the synthesis of aluminophosphates and vanadium aluminophosphates.
The effect of
variation in synthesis parameters, namely template content, Al/P ratio, water content, heteroatom addition, pH and time on the rate of crystallization are discussed. The crystallization kinetics of AlPO4-5 is reported in detail. The next chapter (IV) presents the results of the characterization of AlPO4-5, AlPO4-22, AlPO4-16, AlPO4-L, VAPO-5, VAPSO-5, VAPO-31 and VAPO-41 molecular sieves by different physicochemical methods. Elemental analysis gives the chemical composition of the synthesized materials. X-ray diffraction measurements are used to identify the phase purity of the synthesized materials and their stabilities on calcination. Phase purity and extent of crystallinity are also determined by Infrared spectroscopy, scanning electron microscopy and adsorption methods.
The thermal properties of
occluded templates are analyzed by TGA/DTA and TG-MS techniques.
The
incorporation of vanadium, its co-ordination and oxidation state in as-synthesized and calcined molecular sieves are assessed by electron spin resonance spectroscopy, UVVisible spectroscopy and cyclic voltammetric studies.
Finally, the incorporation of
silicon in AlPO4-35 and VAPO-5 frameworks is examined by
29
Si NMR and FT-IR
spectroscopies. Chapter-V presents the details of the various reactions carried out over the aluminophosphate isomorphism.
An acid catalyzed acetal formation reaction using
benzaldehyde is reported on over AFI type catalysts. Partial oxidation of toluene and cylcohexene are studied over VAPO-5, VAPSO-5 and VAPO-31 molecular sieves. Electrocatalysis of toluene oxidation using VS-1 modified electrode is attempted. Further the oxyfunctionalization of n-hexane over VAPO-31 and VAPO-41 is reported. Finally, a summary of the above studies is presented as chapter VI.
23
2.1 : REFERENCES 1.
Breck, D.W., ‗Zeolite molecular sieves‘, Wiley, New York, (1974).
2.
Wilson, S.T., Lok, B.M., and Flanigen, E.M., U.S. Pat., 4,310,440 (1982).
3.
Wilson, S.T., Lok, B.M., Messina, C.A., Cannon, T.R., and Flanigen, E.M., J. Am. Chem. Soc., 104, 1146 (1982).
4.
Lok, B.M., Messina, C.A., Patton, R.L., Gajek, R.T., Cannon, T.R., and Flanigen, E.M., U.S. Pat., 4,440,871 (1984).
5.
Messina, C.A., Lok, B.M., and Flanigen, E.M., U.S. Pat. 4,544,143 (1985).
6.
Wilson, S.T. and Flanigen, E.M., U.S. Pat., 4,544,143 (1985).
7.
Flanigen, E.M., Lok, B.M., Patton, R.L., and Wilson, S.T., Pure and Applied Chem., 58, 1351 (1986).
8.
Wilson, S.T., Lok, B.M., Messina, C.A., Cannon, T.R., and Flanigen, E.M., A.C.S. Symp. Ser., 79, 218 (1983).
9.
Shannon, R.D., Acta Cryst., 32A, 751 (1976).
10.
Dietzel, A., Poetel, H.J., Die, Naturwissonschaffen, 40, 604 (1953).
11.
Kuiep, R., Mootz, D., Acta cryst., B29, 2292 (1973).
12.
Kuiep, R., Mootz, D., Vegas, A., Acta cryst., B29, 2292 (1973).
13.
Lok, B.M., Messina, C.A., Patton, R.L., Gajek, R.T., Cannon, T.R., and Flanigen, E.M., J. Am. Chem.. Soc., 106, 6092 (1984).
14.
Pyke, D.R., Whitney, P., and Houghton, H., Appl. Cat., 18, 173 (1985).
15.
Kamble, K.R., Kulkarni, S.B., Joshi, P.N., Hegde, S.G., and Ratnasamy, P., Proc. Adv. Cat. And Cat. React. Eng. Ed. P. Kantarao, p. 129 (1986).
16.
Lok, B.M., Marcus, M.T., Kristofforse, B., and Flanigen, E.M., U.S. Pat., 604,155 (1984).
17.
Lok, B.M., Marcus, B.K., and Flanigen, E.M., Eur. Pat. Appl., EP 161,488; 161,489; 161,490 (1985).
18.
Lok, B.M., Vail, L.D., and Flanigen, E.M., Eur. Pat. Appl., EP 158,976; 158,977; 158,348; 158,349; and 159,624 (1985).
19.
Lok, B.M., Marcus, B.K., Vail, L.D., Flanigen, E.M., Patton, R.C., and Wilson, S.T., Eur. Pat. Appl., Appl. EP 159,624 (1985).
20.
Lok, B.M., Marcus, B.K., and Flanigen, E.M., U.S. Pat., 4,500,651 (1985).
21.
Flanigen, E.M., Lok, B.M., Patton, R.L., and Wilson, S.T., Stud. Surf. Sci. Catal., 28, 103 (1986).
22.
Lok, B.M., Marcus, B.K., and Flanigen, E.M., Eur. Pat. Appl., EP 158,350 (1985).
23.
Chao, K.J., Tasi, T.C., Chen and Warg, I.J., J.Chem. Soc., Faraday Trans., I, 77, 547 (1981).
24.
Denpsey, E., Kuhl, G.H.,and Olson, D.H.J., Phys. Chem. 387 (1969).
25.
Smith, J.V., Acta cryst., 15, 835 (1962).
26.
Flanigen, E.M., Khatami, H., Symanski, H.A., ‗Molecular sieve zeolites‘ Adv. Chem. Ser. 101, Amer. Chem. Soc., Washington, DC, p. 201 (1971).
27.
Lohodny-Sorac, D., and White, J.L., J.Phys. Chem., 75, 1408 (1971).
28.
Whyte, J.R., and Dalla Betta, R.A., Cat. Rev. Sci. Eng., 24, 580 (1982).
29.
Ward, J.W., ‗Zeolite chemistry and catalysis‘ Ch. 2, ed. Rabo, J.A., ACS monograph, 171 (1976).
30.
Bond, G.C., Gelsthorpe, M.C., Kenneth, S.W., Sine and Theocharis, C.S., J.Chem. Soc., Chem. Commun., 1050 (1985).
31.
Hegde, S.G., Kustov, L.M., Ratnasamy, P., and Kazansky, V.B., Zeolites, 8, 137 (1988).
32.
Derouane, E.G., Detremmerie, S., Gabelica, Z., and Blom, N., Appl. Catal., 1, 20 (1981).
33.
Barrer, R.M., and Langloy, D.A., J. Chem. Soc., 1817, 3804, 3811 (1958).
24
34.
Ga, I.G., Tankovic, O., Makis, S., Raoovanov, P., and Tadorovic, M., Trans. Faraday Soc., 67, 999 (1971).
35.
Bremen, H., Morke, W., Schodel, R., Vogt, F., Adv. In Chem. Ser. 121, 249 (1973).
36.
Anderson, J.R., Mole, F.K., Rajadiyaksha, R.A., and Sanders, J.V., J. Cat., 58, 114 (1979).
37.
Venuto, P.B., and Landis, P.S., Adv. In. Cat., 18, 269 (1980).
38.
Olson, D.H., Kokatailo, G.T., Lowton, S.L., and Meiser, W.M., J.Phys. Chem. 85, 2238 (1981).
39.
Brunauer, S., ‗The adsorption of gases and vapours‘ Oxford Univ. Press, London, p. 150 (1944).
40.
Derouane, E.G., and Gabelica, Z., J. Catl, 65, 486 (1980).
41.
Gabelica, Z., Gilson, J.P., and Derouane, E.G., ‗Proc. Of 2nd Erup. Symp. On Thermal Catalysis‘, Ed. David Dollimore, Heyden, 434 (1981).
42.
Hong-Xin Li, Martins, J.A., Jacobs, P.A., ‗Innovation in Zeolite Material Science‘, Ed., Gribet, P.J., p. 75 (1987).
43.
Tapp, N.J., Milestone, N.B., and Biby, D.M., ibid, p. 393.
44.
Whittington, B.I., and Anderson, J.R., J.Phys. Chem. 95, 3306 (1991).
45.
Ratnasamy, P., and Kumar, R., Catal. Today, 9, 329 (1991).
46.
Kucherov, A.V., and Slinkin, A.A., Zeolites, 7,43 (1987).
47.
Kucherov, A.V. and Slinkin, A.A., Zeolites, 8, 110 (1988).
48.
Fajes, F., Marsi, I., Kirisci, I., Halasz, J., Hannus, H., Rockenbaker, A., Gy Tasi, Korecz, L., and Schobel, Gy., Stud. Surf. Sci. Catal., 69, 173 (1991).
49.
Kornatowski, J., Sychev, M., Goncharuk, V., and Baur, W.H., Stud. Surf. Sci. Catal., 65, 581 (1990).
50.
Rigutto, M.S., and Van Bekkum, H., Appl. Catal., 68, L1 (1991).
51.
Kornatowski, J., Wichter lova, B., Jirkovsky, J., Loffler, E., and Pilz, W., J. Chem. Soc. Far. Trans., 92, 1067 (1996).
52.
Lippmaa, E., Magi, M., Samoson, A., Engelhardt, A., and Grimmer, A.R., J. Am. Chem.Soc., 102, 4889 (1980).
53.
Nagy, J.B., and Derouane, A.E.G., ACS symp. Ser., 368, 2 (1988).
54.
Lippmaa, E., Magi, M., Samoson, A., Tarmak, M., and Engelhardt, G., J. Am. Chem. Sic., 103, 4992 (1981).
55.
Fyte, L.A., Gobbi, G.C., Klinowski, J., Thomas, J.M., and Ramdas, S., Nature, 296, 530 (1982).
56.
Fyfe, C.A., Gobbi, G.C., Kilinowski, J., Thomas, J.M., J. Phys. Chem., 86, 1247 (1982).
57.
Engelhardt, G., Lohose, U., Lippmaa, E., Tarmak, M., and Magi, M.Z., Anorg. Allg. Chem. 482, 49 (1981).
58.
Mastikhin, V.M., and Zmaraev, K.I.Z., Phys. Chemie, 152,332 (1987).
59.
Muller, D., Gassenev, W., Behrens, H.J., and Scheler, G., Chem. Phys. Lett., 79, 59 (1981).
60.
Thomas, J.M., Fype, C.A., Ramdas, S., Kilinowski,J., and Gobbi, G.C., J.Phys. Chem., 86, 3061 (1982).
61.
Ramdas, S., and Klinowski, J., Nature, 308, 521 (1984).
62.
Barrer, R.M., and Ibbitson, D.A., Trans. Faraday Soc., 40, 206 (1944).
63.
Castro-Martins, S.D., Khouzanpi, S., Tuel, A., and Taarit, Y.B., Mur, N.E., and Sellami, A., J. Electroanal. Chem.m 350, 15 (1993).
64.
Castro-Martins, S.D., Tuel, A., and Taarit, Y.B., Zeolites, 14, 130 (1994).
65.
Gortsema, F.P., Springer, A.R., Long, G.N., U.S. Pat., 4,880,700 UOP (1989).
66.
Pellet, R.J., Coughlin, P.K., Long, G.N., and Rabo, J.A., U.S. Pat., 4,859,314 UOP (1989).
25
67.
Pellet, R.J., Coughlin, P.K., Stainiulis, M.T., Rabo, J.A., and Long, G.N., U.S. Pat., 4,791,083, UOP (1988).
68.
Gortsema, F.P., Pellet, K.J., Springer, A.R., Rabo, J.A., and Long, G.N., U.S. Pat., 4,913,799 UOP (1990).
69.
Gortsema, F.P., Springer, A.R., Rabo, J.A., and Long, G.N., U.S.Pat., 4,818,739 UOP (1989).
70.
Bennett, J.M., and Kirchner, R.M., Zeolites, 11, 503 (1991).
71.
Richardson Jr., J.W., Pluth, J.J., and Smith, J.V., Naturwissenschaften, 76, 467 (1989).
72.
Barrer, R.M., and Kerr, I.S., Trans. Farad. Soc., 55, 1915 (1959).
73.
Bennet, J.M., and Kirchner, R., Zeolites, 12, 338 (1992).
74.
Bennet, R.M., NATO Advanced Study Institute, Alcabideche, Portugal, May (1991), (private communication).
75.
Bialek, R., Meier, W.M., Davis, M.,and Annen, M.J., Zeolites, 11, 438 (1991).
76.
In ―Atlas of Zeolite Structures‖, Ed., Meieer, W.M., and Olson, D.H., Zeolites, 12, 16 (1992).
77.
Huckneall, D.J., Selective oxidation of organic compounds, Academic Press, New York (1974).
78.
Dadyburyor, D.B., Jewur, S.S., and Rukenstein, E., Catal. Rev. Sci. Eng., 19, 293 (1979).
79.
Kornatowski, J., Sychev, M., Goncharuk, V., and Baur, W.H., Stud. Surf. Sci. Catal., 65, 581 (1991).
80.
Rigutto, M.S., and Van Bekkum, H., Appl. Catal., 68, L1 (1991).
81.
Rao, P.R.H., ramaswamy, A.V., and Ratnasamy, P., J. Catal., 137, 225 (1992).
82.
Tuel, A., and Ben Taarit, Y., Appl. Catal. A: General, 102, 201 (1993).
83.
Reddy, K.M., Moudrakovski, I., and Sayari, A., J. Chem. Soc., Chem. Commun., 1059 (1994).
84.
Reddy, K.M., Moudrakovski, I., and Sayari, A., J. Chem. Soc., Chem.Commun., 1491 (1994).
85.
Hausen, D.M., Am. Mineral, 47, 637 (1962).
86.
Flanigen, E.M., Lok, B.M.T., Patton, R.L., Wilson, S.T., Eur. Pat. Appl., 0,158,976 (1985).
87.
Pyke, D.R., Whitney, P., and Houghton, H., Appl. Catal., 18, 173 (1985).
88.
Miyamoto, A., Iwamoto, Y., Matsuda, H., and Inui, T., Studies in Surf. Sci. Catal., 49B, 1233 (1989).
89.
Montes, C., Davis, M.E., Murray, B.M., and Narayana, M., J.Phys. Chem. , 94, 6431 (1990).
90.
Jhung, S., Uh, Y.S., and Chon, H., Applied Catalysis, 62, 61 (1990).
91.
Kornatowski, J., Sychev, M., Baur, W.H., and Finger, G., Collect. Czech. Chem. Commun., 57, 767 (1992).
92.
Haanepan, M.J., and van Hooff, J.V.C., in DGMK conference, ―Selective oxidations in Petrochemistry‖, on September 10-18, Goslar, Germany, p. 227 (1992).
93.
Rigutto, M.S., and van Bekkum, H., J. Mol. Catal., 81, 77 (1993).
94.
Blasco, T., Concepcion, P., Lopez Nieto, J.M., and Perez-Pariente, J., J. Catal., 152, 1 (1995).
95.
Weckhuysen, B.M., Vannijvel, I., and Schoonheydt., R.A., Zeolites, 15, 482 (1995).
96.
Miller, S.J., U.S. Pat., 4,859,311 (1989).
97.
Blackwell, C.S., and Patton, R.L., J.Phys. Chem., 92, 3965 (1988).
98.
Zubowa, H., Alsdort, E., Fricke, R., Neissendorfer, F., Richter-Mendaeu, J., Schreier, E., Zeigan, D., and Zibrowins, B., J.Chem.Soc. Far. Trans., 92, 2307
26
(1990). 99.
Singh, P.S., Bandyopadhyay, R., and Rao, B.S., J. Chem. Soc. Far. Trans., 92, 2017 (1996).
100.
Vomscheid, R., Briend, M., Peltre, M.J., Man, P.P.,and Barthomeuf, D., J. Phys. Chem., 98, 9614 (1995).
101.
Herreros, B., and Klinowski, J. Phys. Chem., 99, 9514 (1995).
102.
Qisheng, H., Ruren, X., J. Chem. Soc. Chem. Commun., 783 (1990).
103.
Gao, Q., Li, S., and Xu, R., J. Chem. Soc., Chem. Commun., 1465 (1994).
104.
Lohse, U., Vogt, F., Richter-Mendau, J. Cryst. Res. Technol., 28, 1101 (1993).
105.
Barrett, P.A., Jones, R.H., Thomas, J.M., Sanker, G., Shannon, I.J., Catlow, C.R.A., J. Chem. Soc., Chem. Commun., 2001 (1996).
106.
Rubin, M.K., and Chu, P., U.S. Pat., 4,954,325 (1990).
107.
Puppe, L., and Weisser, J., U.S. Pat., 4,439,409 (1984).
108.
Marler, B., Dehnbostel, N., Eulert, H.H., Gies, H., and Filiebau, J. Include. Phenom., 4, 339 (1980).
109.
Liebau, F., ―Structural Chemistry of Silicates, Structure, bonding and Classification‖, Springer-Verlag, Berlin, p. 240 (1985).
110.
Huo, Q., Xu, R., Li, S., Na, Z., Thomas, J.M., Jones, R.H., and Chippindale, A.M., J. Chem. Soc. Chem.Commun., 875 (1990).
27
Table 1.1: Classification of aluminophosphate molecular sieves according to the size of pore openings. Pore size and name
structure Pore size (nm)
Small pore (8MR)* AlPO4-16 AST AlPO4-20 Sodalite AlPO4-25 ATV AlPO4-28 Novel AlPO4-14 Novel AlPO4-17 Erionite AlPO4-18 AEI AlPO4-26 Novel AlPO4-33 ATT AlPO4-34 Cehabaz ite AlPO4-35 Levynite AlPO4-39 ATN AlPO4-42 Linde type A AlPO4-43 Gismon dine AlPO4-44 Chabazit e AlPO4-47 Chabazit e
Pore volume (H2O) cc/g
0.30 0.30 0.30 0.30 0.40 0.43 0.43 0.43 0.40 0.43
0.34 0.24 0.17 0.21 0.19 0.28 0.35 0.23 0.23 0.30
0.43 0.40 0.43
0.30 0.23 0.30
0.43
0.30
0.43
0.34
0.43
0.30
Pore size and name
structure
Pore size (nm)
Medium pore (10MR)* AlPO4- 11 AEL 0.60 AlPO4-31 ATO 0.65 AlPO4-41 AFO 0.60 Large pore (12MR)* AlPO4-36 ATS 0.80 AlPO4-37 Faujasite 0.80 AlPO4-46 Novel 0.70 Extra large pore (>12MR)* VPI-5 VFI AlPO4-8 AEI
*
- Type of ring in pore
28
Pore volume (H2O) cc/g 0.16 0.17 0.22 0.35 0.33 0.28
29
CHAPTER – II ________________________________________________________________________ EXPERIMENTAL DETAILS ________________________________________________________________________
30
2.1: INTRODUCTION The aluminophosphate molecular sieves have been synthesized using reactive sources of alumina, orthophosphoric acid, templating agent and water and heating the gel under hydrothermal conditions at autogeneous pressure. The sources of raw materials and the methods used for their chemical analysis are discussed briefly in this section. The different instrumental techniques employed in elucidating the structural features and properties of the synthesized products are also described. 2.2: SYNTHESIS AlPO4 – molecular sieves were synthesized hydro thermally using the reactants presented in Table 2.1. In a typical procedure, an aluminium source (pseudoboehmite or aluminium isopropoxide) was mixed with a solvent (water of ethylene glycol) thoroughly. Orthophosphoric acid was added drop wise to the mixture and stirred well to obtain an aluminophosphate gel. To this gel, the required template was added and stirred well to make the final gel.
The final gel composition and conditions were varied
depending on the structure. A vanadium source (V2O5 or VOSO4.xH2O) and or a silicon source (fumed silica or silica sol) were added during the gel formation to obtain VAPO or SAPO or VAPSO molecular sieves. Crystallization was carried out using a Teflon lined stainless steel autoclave (Fig. 2.1) at 423-473 K for 1-15 days at autogeneous pressure. The final product was filtered, washed with distilled water several times and dried at ambient temperature. The gel molar composition and duration of crystallization used for various AlPO4 and VAPO molecular sieve are given in Table 2.2. 2.3: PRETREATMENT PROCEDURE All the aluminophosphate synthesized above contained organic templates inside the channels. In order to characterize and use these aluminophosphates in catalytic reactions, it was necessary to remove the organic material from the pores of the aluminophosphates. The as-synthesized material was placed in a Petri-dish and calcined in a flow of nitrogen at 673 K for 8h followed by passing dry air and raising the temperature to 823 K (12h). The temperature of the furnace was raised at a rate of 2.5 K/min. In order to remove the polymeric and nonlattice vanadium present in the calcined vanadium aluminoposphates, the calcined samples were refluxed in 1M ammonium acetate solution (20 ml/g, 373 K for 8 h). After the treatment the sample was filtered washed thoroughly with deionized water and dried at 393K for 10 h. The resultant material was calcined 723 K. 2.4: METHODS USED FOR PHYSICOCHEMICAL CHARACTEIZATION 2.4.1: Chemical analysis 2.4.1A: Estimation of aluminium as 8-hydroxy quinolate [Al(C9H6OH)3] A known quantity of the aluminophosphate sample was dissolved in a few ml of con. HCl, diluted and heated 343-353 K and an appropriate amount (about 40 ml) of the oxime reagent was mixed and a 2M ammonium acetate solution was then added slowly until a precipitate just appeared. The mixture was heated to boiling and again 25 ml of 2M ammonium acetate solution was added drop wise and with constant stirring. The precipitate was allowed to cool and filtered through a quantitative filter paper (whatman no. 41) and washed with ice cold water. The precipitate was transferred into a previously weighed platinum crucible, dried, charred, ignited and weighed as Al2O3. 2.4.1B: Estimation of phosphorous as magnesium phosphate (Mg2P2O7) A known amount of the sample (about 0.5 g) was dissolved in a few ml of con. HCl, diluted and boiled. A few drops of methyl red indicator were added to keep the aluminium in solution. Then 25ml of magnesia mixture was added followed by pure concentrated ammonia solution while stirring vigorously until the indicator turned yellow.
The solution was allowed to stand at room temperature for 24 hrs.
The
precipitate was then filtered, washed with dilute ammonia solution until the filtrate gave no turbidity with AgNO3 and HNO3. The precipitate was transferred to a platinum crucible, dried, charred and ignited to a constant weight and weighed as Mg2P2O7. 31
2.4.1C: Estimation of silicon About 1g of the sample was ignited in a platinum crucible to a constant weight. After cooling it was moistened with a few drops of con. H2SO4 and 5 ml of AR hydrofluoric acid was added and slowly evaporated to dryness. The HF treatment was repeated thrice or more till there was no further weight loss. From the loss in weight, silica was estimated. 2.4.1D: Estimation of carbon and nitrogen The carbon and nitrogen contents in the as-synthesized samples were analyzed by a CHNS-O elemental analyzer (CARLO ERBA, model EA1108). 2.4.1E: Estimation of V4+ and V5+ ions 1g of the catalyst was dissolved in con. H2SO4 or HF and made up to 250 ml solution. 100 ml of the solution was heated at 333 K and then titrated with 0.1 M KMnO4 solution to determine V4+ and then after cooling it was titrated with 0.1 N ferrous ammonium sulfate solution using sodium diphenylamine (C6H5NHC6H4SO3Na) as an indicator (sharp color change from violet to green) to determine the total vanadium content. Another lot of 100ml of the solution was directly titrated with Fe2+ solution to determine the V5+ content. The total vanadium concentrations estimated by the titrimetric method matched the values obtained by AAS (Hitachi) and ICP (Jobin Yvon-JY-38) analysis. To avoid the transformation of V4+ and V5+ during the dissolution of the catalyst prior to titration, the dissolution of the solid catalysts was carried out under O2 – free CO2 and the titration was performed in a CO2 atmosphere. The titrimetric procedure was also standardized by the analysis of a series of silicalite -2 samples impregnated with known quantities of V4+ and V5+ ions. 2.4.1F: Ion exchange About 2g of the calcined sample was equilibrated with 50 ml of 0.1 M NaCl solution at 333 K for 30 min. The sample was filtered and washed with deionised water till the filtrate was free from Na+. The sample was dried at 373 K for 12 hours. About 0.5 g of the above dried sample was dissolved in a few ml of con. HCl or H2SO4 and diluted to 250 ml.
The solution was analyzed for sodium by atomic absorption
spectroscopy while aluminium and phosphorous estimated according to the procedure described earlier. 2.4.2: X-ray diffraction The samples synthesized during the course of the work under different conditions and different crystallization times were analyzed for qualitative and quantitative phase identification by X-ray powder diffraction (Rigaku, Model D/MAX III VC, Japan; Ni filtered radiation, lambda = 1.5404 Ao; controlled automated diffractometer).
graphite crystal monochrometer; computer
XRD was also used to identify the different
phases. To Identify the lattice structure of the synthesized unknown phase, the sample was ground and equilibrated with air for 24 hours and scanned in the 2 theta range of 540o with a step size of 0.02o and counting time of 10 sec at each step. The sample holder was rotated throughout the scan. Silicon was used as the internal standard. The relative intensities of the observed peaks were obtained after smoothing, background correction and Kalpha2 stripping. The observed inter planar distance ‗d‘ of the observed peaks were corrected with respect to silicon. The corrected inter planar ‗d‘ spacing were indexed for different lattice structures using an automatic indexing program (PDP11 and HOCT) and checked with trial and error methods. The unit cell dimensions were refined using a least square fitting. The X-ray crystallinity was calculated for AlPO4-5 from the following equation. The most crystalline sample synthesized during the experimental work was taken as the standard.
32
sum of the peak intensities at 2theta (degree) values of X-ray crystallinity = 7.4, 14.92, 19.70, 21.14, 22.38 and 25.94 of AlPO4-5
x 100
Sum of the peak intensities at 2 theta (degree) values of 7.4, 14.92, 19.70,21.14, 22.38 and 25.94 0f AlPO4-5 2.4.3: Scanning electron microscopy The morphologies of all the aluminophosphates synthesized were investigated using a scanning electron microscope (JEOL, JSM 5200). The sample was dusted on the holder, and coated with a thin film of gold to prevent surface charging and to protect the zeolite material from thermal damage by the electron beam. In all the analysis, a uniform film thickness of about 0.1 mm was maintained. 2.4.4: Infrared spectroscopy The framework region (400-1300 cm-1) of the synthesized aluminophosphates were analyzed using a Nicolet 60SXB FT-IR instrument in the diffuse reflectance mode using a 1:20 ratio of the sample to KBr mixture. For the range 1300-4000 cm-1, self supported wafers were used. For insitu studies, a cell which could be heated to 773 K was fabricated and used (Fig. 2.2).
The cell was connected to a volumetric adsorption
apparatus (Micrometrics, USA, model; accusorb – 2000E). The sample was pressed under a pressure of 5 ton/inch2 into a thin pellet (5-6 mgs/cm2) and mounted in the sample holder (SH). It was then placed inside the heating compartment (HC) of the cell and aligned in line with the IR beam. The cell was sealed at both the ends by potassium bromide windows (KW) with the help of elastomer ―O‖ rings. A thermocouple placed in close vicinity of the sample measured the temperature of the sample. The side tube at the top of the cell was connected to the high vacuum system. Cold water was constantly circulated through the cooling coil provided near the KBr windows. The sample was heated to 673 K at a heating rate of 5 K/min under vacuum and maintained at this temperature for about 6 h and then cooled down to 323 K in vacuum. The spectra were recorded by averaging more than 100 scans with 2 cm-1 resolution. Vapours of ammonia were admitted to the sample through the adsorption manifold of the system. 2.4.5: Electron spin resonance spectroscopy Electron spin resonance spectra of the as-synthesized and calcined vanadium aluminophosphates were obtained using Bruker ER 200D spectrometer. The spectra were recorded at room temperature as well as at liquid nitrogen temperature. The spectra were recorded (with a mid range of 3500 G) in the scan range of 2000 to 5000 gauss. The magnetic field strength was 9.74 GHz. 2.4.6: UV-Vis spectroscopy The UV-Vis diffuse reflectance spectra were recorded using a Pye Unichem SP-8100 UV-Visible spectrometer I the 200-900 nm region. 2.4.7: Thermal analysis Simultaneous TG/DTA analysis of the crystalline phases was performed on an automatic derivatograph (Setaram, TG-DTA 92). The thermograms for the samples were recorded under the following conditions. Weight of the sample = 30 mg Heating rate = 10 K/min. Atmosphere = flowing air or argon Preheated and finely powdered alpha alumina was used as reference material. Thermal desorption mass spectroscopic analysis was carried out using a commercial instrument (Sorbstar model 100) supplied by the Institute of Isotopes, Budapest, Hungary. The temperature range studied was from room temperature to 805 K; argon was used as the carrier gas. 2.4.8: Nuclear magnetic resonance spectroscopy MASNMRspectra were recorded in the solid state with a Bruker MSL 300 spectrometer operating at a field of 7 Tesla. 27Al spectra were recorded at a frequency of 78.2 KHz, with a pulse length of 2us and a spinning speed of 3-5 KHz. 33
13
C CP
MASNMR were recorded at 75.47 MHz using CP pulse spectrum. The Hartman Hann match conditions were established using adamantine (typically a match condition of 50 KHz was employed). The chemical shifts were referred to with respect to the C-H peak of adamantine (38.7 ppm). A spinning speed of ~3 KHz was used. The high resolution spectra of 29 Si were obtained at room temperature using ―one cycle‖ type measurements. A 90o pulse with 3 us delay time was used for
29
Si nuclei and chemical shifts (delta) in
ppm were measured with respect to tetramethyl silane (TMS) as the external reference. The rotor (sample holder was spun at a rate of 4.0 KHz. 2.4.9: Cyclic voltammetry The modified electrodes with molecular sieves were prepared by coating a small amount (ca. 6 mg) of a paste made-up of 100 mg of the molecular sieve, 100 mg of pure graphite and 10 mg of polystyrene in 2 cm3tetrahydrofuran on the tip of a platinum ultra microelectrode (UME). The details regarding different chemicals used in the study are presented in Table 2.3. In some cases a platinum micro electrode (6 mm diameter) was used. The UME (10 um diameter) was polished with diamond paste before being coated with the molecular sieve paste. The paste was then allowed to dry under vacuum. Cyclic voltammetric experiments were carried out using a three electrode single compartment cell (Fig. 2.3) with the molecular sieve coated Pt UME as the working electrode, platinum foil as the counter electrode and standard calomel electrode (SCE) as the reference electrode in 0.1 M aqueous KCl (pH = 5.24). The experiments were carried out using a EG&G Princeton Applied Research potentiostat (PAR 173) coupled with a function generator (PAR173), together with an X-Y recorder (model RE 0091). Background voltammograms were recorded with modified electrodes having molecular sieves with out any electroactive species in the region +1.2 to -0.5 V. The electrolyte solution was deoxygenated before all measurements were taken by purging with argon. All the experiments were carried out at a constant temperature of 298+1 K. Controlled potential coulometric (CPC) experiments were carried out at the potential -0.0125 and 0.125 V vs SCE using VS-1/graphite, graphite and silicalite/graphite pellets (diameter, 6 mm) in 0.1 mol dm-3 KCl solution under an argon atmosphere. The coulometric products were analyzed using XRD, FT-IR and EPR. 2.4.10: Catalytic properties Details regarding different chemicals used in the catalytic reactions are presented in Table 2.3. 2.4.10 A: n-hexane oxyfunctionalization The catalytic reactions were carried out in stirred autoclaves (Parr Instrument company, Illinois, USA) of 300 ml capacity at a temperature of 393 K under autogeneous pressure with an alkane : H2O2 mole ratio of 3. In order to get a single phase, acetone was used as the solvent. Typically, 0.5 g of the catalyst was used in the reaction. The duration of the run was 8h. 2.4.10 B: Benzaldehyde acetalization 0.52 g of benzaldehyde was mixed with 10 ml of methanol in a round bottom flask (50 ml).
40 mg of catalyst was then added and the reaction mixture stirred
continuously at room temperature for 24 h. The product was removed at different time intervals and analyzed. 2.4.10 C: Toluene and cyclohexene oxidation The reactions were carried out in a two necked round bottom flask. Tertiary butyl hydro peroxide and the reactant were taken in an equimolar ratio along with ten fold excess of the solvent chlorobenzene. The reaction was carried out at 373 K for 22 h for toluene oxidation and for 6 h for cyclohexene epoxidation. 2.4.10 D: Electrocatalysis VS-1 modified graphite paste pellet (6mm diameter) was used as the working electrode, standard calomel electrode as the reference electrode and platinum foil as the auxiliary electrode. These were fixed in a 100 ml beaker kept over a magnetic stirrer. 34
Equal molar ratios of toluene and hydro peroxide were taken along with a ten fold weight of acetonitrile (solvent). The reaction mixture was kept at a room temperature and stirred continuously. A potential of 1 V was applied on the working electrode continuously for 12 h. The current (coulombs) consumed in the oxidation was measured. The reaction product was analyzed in a Gas chromatograph. The above reaction was also carried out separately at 353 K for 12 h in a round bottom flask using the same molar composition of reactants for comparison purpose. 2.4.10 E: Analysis The products of all the reactions were analyzed in a gas chromatograph (HP 5880A) using a capillary column (50 m x 0.5 mm; HP1 cross linked methyl silicone gum) and FID.
Standard mixtures were used for calibration purposes, while component
identification was carried out separately using a GC-MS (Shimadzu, model QP2000A).
35
Fig. 2.1: Stainless steel autoclave with Teflon gasket and liner for hydrothermal synthesis.
36
Fig. 2.2: Sample cell for FT-IR spectrometer
37
Fig. 2.3: Block diagram of cyclic voltammetry
38
Table 2.1: Specification of the reactants used in the synthesis S. No. 1 2
Chemical Name Aluminium isopropoxide (Aldrich) Pseudoboehmite (Vista)
Chemical Formula Al(OC3H7)3 Al2O3
3 4
Fumed silica (Fluka) Orthophosphoric acid (s.d. fine, India)
SiO2 H3PO4
5 6 7 8 9 10 11 12 13
Hexamethyleneimine (Aldrich) Dipropylamine (Aldrich) Diethyl amine (Loba, India) Dibutyl amine (Aldrich) Dipentyl amine (Aldrich) Tripropyl amine (Aldrich) Vanadium pentoxide (Aldrich) Vanadyl sulphate hydrate (Aldrich) Vanadyl sulphate penta hydrate (Loba, India).
C6H12NH (C3H7)2NH (C2H5)2NH (C4H9)2NH (C5H11)2NH (C3H7)3N V2O5 VOSO4xH2O VOSO4XH2O
39
Purity (%) 98 74.2 (25 %, H2O) 99.5 85 (15 %, H2O) 99 99 98 99 99 98 98 100 95
Table 2.2: Gel molar compositions and duration of crystallization for various AlPO4 and VAPO molecular sieves. Structure AlPO4-5 AlPO4-22 AlPO4-16 SAPO-35 AlPO4-L* VAPO-5 VAPSO-5 VAPO-31 VAPO-41
Gel composition
Duration of crystallization Al2O3: P2O5: 1.16HEM: 45H2O 24h Al2O3: 1.15P2O5: 1.16HEM: 45H2O 24h Al2O3: P2O5: 1.35HEM: 45H2O 24h Al2O3:1.8 P2O5: 4.5HEM: 0.3SiO2: 15d 45EG Al2O3: P2O5: 2.32HEM: 45H2O 24h Al2O3: P2O5: 1.16HEM:0.010.08V2O5: 45H2O Al2O3: P2O5: 1.16HEM: 0.08V2O5: 0.2-1.0SiO2: 45H2O Al2O3: P2O5: 1.16DPA: 0.020.05V2O5: 45H2O Al2O3: P2O5: 1.16DEA: 0.02V2O5: 45H2O
Other variables
24h
Si-source: fumed silica Al-source: CatapalB or aluminium isopropoxide -
24h
-
48h
V-source: V2O5
5d
V-source: VOSO4.5H2O
Conditions: 473 K, autogeneous pressure. *
- a novel lamellar type aluminophosphate, HEM = hexamethyleeimine, EG = ethylene glycol, Al-source = Catapal – B, P-source = orthophosphoric acid, silicon source = silica sol, Vanadium source = VOSO4.xH2O.
40
Table 2.3: The reagents used in cyclic voltammetric experiments and various catalytic reactions. S.No. 1 2 3 4 5 6 7 8 9 10 11 12 13
Chemical Potassium chloride (Aldrich) Tetrabutyl ammonium tetra fluo borate (Aldrich) Lithium perchlorate (Aldrich) Hydrogen peroxide (s.d.fine, India) Tert-butyl hydroperoxide (Aldrich, USA) n-hexane (s.d.fine, India) Cyclohexene (s.d.fine, India) Toluene (s.d.fine, India) Benzaldehyde (s.d.fine, India) Methanol (s.d.fine, India) Acetonitrile (s.d.fine, India) Acetone (s.d.fine, India) Chloro benzene (s.d.fine, India)
Formula KCl Bu4NBF4
Purity (%) 98 99
LiClO4 H2O2
100 30 w/v (in H2O)
(CH3)3COOH
70 wt% (in H2O)
C6H12 C6H10 C7H14 C6H5CHO CH3OH CH3NC CH3COCH3 C6H5Cl
99 99 99 99 99 99 99 99
41
42
CHAPTER III ________________________________________________________________________ SYNTHESIS ________________________________________________________________________
43
3.1: INTRODUCTION In this chapter, the hydrothermal synthesis of aluminophosphate molecular sieves using hexamethyleneimine template is discussed. During the study, as a result of variation in template, concentration, Al:P ratio, solvent concentration, hetero atoms and temperature, various structure types of molecular sieves like AlPO4-5, AlPO4-22, AlPO416, SAPO-35, VAPSO-5, VAPO-31 and VAPO-41 are obtained. 3.2: SYNTHESIS OF AlPO4 AND VAPO MOLECULAR SIEVES 3.2.1: Variation in template concentration The reactive gel composition, Al2O3: P2O5: 1.16HEM: 45H2O was standardized after several initial experiments and used as the basis for further study.
The
concentration of HEM was varied from 1.16 moles to 2.32 mole and the products obtained at different times of crystallization are tabulated in Table 1. It is found that when the HEM content is 1.16 mole rapid crystallization occurs to yield pure AlPO4-5 in 4.5h. This transforms into AlPO4-22 at 192h. When a slightly larger amount of HEM (1.35 moles) is used, AlPO4-16 results after 24h. At still higher HEM levels, a lamellar phase, AlPO4-L is produced [11]. This aluminophosphate (AlPO4-L) phase is observed during the conversion of AlPO4-5 to AlPO4-22 and as an intermediate phase in many other experiments [11]. 3.2.2: Influence of Al/P ratio Generally, aluminophosphate molecular sieves are prepared by using equimolar amounts of Al and P. If the gel composition is varied slightly from the preferred mole ratio, different phases are formed. This behavior was investigated by changing the mole ratios of Al and P as shown in Table 2(i). For the typical gel composition, Al 2O3: P2O5: 1.16HEM: 45H2O, the crystallization period and temperature were maintained at 24h and 473K respectively. When the moles of Al in the gel exceeded P substantially, excess boehmite lead to a mixture of AlPO4-5 and an unidentified impurity. However when P in the gel substantially exceeded Al (0.8 molar ratio), AlPO4-22 was formed. This is believed to be due to the presence of excess phosphorous ions in the AlPO4-22 structure. Excess phosphate typically added as H3PO4 generally has a destructive effect on AlPO4-5 formation due to the low pH of the gel. 3.2.3: Influence of water content With the aim of reducing the crystallization time and improving the time space yield, the cyrstallization was carried out with different water contents. Table 2(ii) shows the crystallization products at various water contents. If the water content (water/P2O5 mole ratio) was lower than 45, AlPO4-16 was formed, and when it was 75 or above AlPO4-22 impurity phases were produced. 3.2.4: Influence of aluminium source The effect of using two different aluminium sources (pseudoboehmite and aluminium isopropoxide) on the crystallization of AlPO4-5 was next examined. The experiments were carried out with the standard reaction mixture: Al2O3:
P2O5:
1.16HEM: 45H2O. In the case of pseudoboehmite, pure AlPO4-5 was obtained, but in the case of aluminium isopropoxide, the product was pure AlPO4-16 even at short crystallization times.
Prolonged crystallization up to two days did not alter the
crystallinity of the phase. The effect was obviously due to the presence of isopropanol in the gel. This was confirmed by experiments conducted with pseudoboehmite gels to which isopropanol was added. This mixture produced only AlPO4-16. 3.2.5: Influence of silica and vanadium oxide addition Isomorphous substitution of silicon and vanadium in AlPO4-molecular sieves generates acidic and redox properties respectively. On addition of silica (as silica sol or fumed silica), the induction period required for crystallization was reduced. Addition of silica to the gel (Al2O3: P2O5: 1.16HEM: 45H2O Al2O3: P2O5: 1.16HEM: 45H2O) up to 0.1M gave an AlPO4-5 type phase and further addition lead to impurity phases (Table 3). There appears to be a critical value above which the silica is no longer incorporated in the 44
lattice, but is precipitated amorphously in the structure affecting AlPO4-5 crystallization. An inspection of the sample by scanning electron microscopy revealed no amorphous particles, but crystallite morphology was altered significantly to yield triangular crystals of 1.5m (Fig. 1a). Addition of vanadium source (VOSO4.xH2O) to the above gel (Al2O3: P2O5: 1.16HEM: 45H2O) gave AlPO4-5 type materials, but the rate of crystallization was considerably decreased. The critical upper limit for vanadium addition was 0.08V2O5 (Table 3); further addition lead to the known dense phases. Morphology of the crystals did not change with vanadium, though the particle size decreased with increasing vanadium content. When tripropylamine was used as the template, the upper critical limit for V-incorporation was found to be 0.06M V2O5.
When Si and V were
simultaneously introduced into the reaction mixture, silicon incorporation increased to above 0.1M SiO2 and no impurity phase was observed up to 1.0M Si in the gel. 3.2.6: Influence of crystallization medium The crystallization of AlPO4-5 using hexamethyleneimine as the template was studied in a nonaqueous medium (ethylene glycol; EG) at 473K using a molar gel of composition Al2O3: 1.8P2O5: 4.5HEM: 45EG. The XRD patterns obtained at different time intervals are presented in Fig. 2. The crystallization kinetics is presented in Fig. 3. The crystallization was very slow in the non-aqueous medium. On the 5th day a highly unstable precursor phase was formed. AlPO4-5 along with some known dense phases (cristobalite and tridymite) was formed on the 10th day. Pure AlPO4-5 was formed on the 15th day. The sample attained maximum crystallinity on the 30th day. The crystallization curve is similar to that in aqueous medium and follows a sigmoid type. Interestingly, the final sample (30th day) was more crystalline (1.4times) than the most crystalline sample prepared in the aqueous medium. The morphology of the sample was hexagonal and the particle size was 4.2m (Fig. 1b). Crystallization of a gel of compositions, Al2O3: 1.8 P2O5: 4.5HEM: 45EG, after aging with stirring for 8h gives AlPO4-31 after 15days at 473K. Incorporation of Si in AlPO4-5 in non-aqueous medium was attempted by adding 0.3M of SiO2 into the reaction gel mixture. Surprisingly, instead of the expected SAPO-5, SAPO-35 was obtained [12]. It may also be noted that SAPO-35 is obtained in aqueous medium when quinuclidine or cyclohexylamine is used as the template [8]. The preparation of SAPO-35 molecular sieve has not been reported todate using other templates. 3.2.7: Influence of pH The role played by the pH of the reaction mixture on the formation of pure AlPO4-5 phase has been examined by adding different quantities of HCl to a typical gel composition. The gel composition was; xHCl: Al2O3: P2O5: 1.16HEM: 45H2O, (x = 01.5). Even though the total concentration of HEM and H3PO4 was kept constant, the addition of HCl produced a mixture of C6H12NH and C6H12NH2+Cl- in the new reaction gel. As a result, HCl had a distinctive effect on the crystallization of AlPO 4-5 (Table 2(iii)). In more acidic media, dense phase‘s viz., cristobalite, tridymite, and berilinite were formed. It is concluded that for a given composition of the gel, its pH should be above 3.0 for the formation of the AlPO4-5 phase. 3.3: Crystallization kinetics of AlPO4-5 The changes in pH during the preparation of the gel (Al2O3: P2O5: 1.16HEM: 45H2O; typical composition) were as follows: The pH of the distilled water (6.11) increased (8.42) on addition of pseudoboehmite due to the partial hydrolysis of alumina. The addition of orthophosphoric acid decreased the pH to 2.03. Dilution by water increased the pH to 2.50. Addition of the templating agent increased the pH further to 4.27. The pH during crystallization varied with temperature and time. Both pH and crystallinities remained constant after attaining a maximum value. The pH of the mother liquor after crystallization was higher due to the free template present in the solution.
45
The course of crystallization of AlPO4-5 as measured from the intensity of the Xray powder diffraction patterns is shown in Fig. 4. The most crystalline material was obtained after 4.5h using a gel of composition Al2O3: P2O5: 1.16HEM: 45H2O, at a reaction temperature of 473K. Also, the influence of crystallization time on crystal morphology was examined by SEM (Fig. 1c-f). AlPO4-5 is stable in its reaction mixture for at least 3 additional days at this temperature. The scanning electron micrographs of the AlPO4-5 crystals synthesized from a gel of the above molar composition revealed spherical cauliflower shaped crystal agglomerates with an average diameter of 8m (Fig. 1f).
The XRD pattern of AlPO4-5 synthesized using triethylamine following the
published procedure [6] had a crystallinity of about 80% (compared to the standard sample synthesized using HEM) and the morphology of the crystals was hexagonal bundles of small needles of 3-5m size. In the typical gel composition described above, crystallization of AlPO4-5 occurs rapidly in the temperature range, 393-473K.
The changes in pH with duration of
crystallization and temperature are presented in Fig. 5.The low pH of the initial gel (pH = 2.7 - 4.5) reflects the presence of some unreacted orthophosphoric acid (H3PO4). When the crystallization is in progress, there is an initial lowering of pH probably due to the dissociation of some acidic molecules during initial realignment. Al and P are consumed in equimolar amounts causing a rise in pH (7 to 9) at higher temperatures during digestion. Final values of the pH in the range 7 to 10 are typical for this gel composition. AlPO4-5 phase is quite stable in the temperature range of 393-473K in the reaction medium. However after reaching maximum crystallinity, a partial dissolution of the crystalline phase occurs at 473K. Prolonged crystallization at 473K leads to AlPO4-22 through an intermediate unstable phase (AlPO4-L).
Increasing the temperature above
473K leads to faster crystallization of AlPO4-22. In this phase transformation, the pH of the reaction medium plays a crucial role. Although AlPO45 can be synthesized at 373K, its crystallinity is poor even after prolonged crystallization. temperatures
and
crystallization
periods
for
obtaining
The most favorable pure
AlPO4-5
using
hexamethyleneimine are in the range of 423-473K and 4 - 36h, respectively. It was observed that at lower crystallization temperature and duration, known hydrated aluminophosphates were formed. AlPO4-22 phase was formed at higher temperatures and on prolonged crystallization. A kinetic study of the crystallization of AlPO4-5 was carried out for selected batch compositions to determine the kinetic parameters, viz., the apparent activation energies for nucleation and crystallization. Pseudoboehmite gave three broad peaks at 2 values of 13.9, 28.2 and 32.3o (Fig. 4a). The template free aluminophosphate gel was composed of a mixture of three dense phases, viz., cristobalite, tridymite and berilinite (Fig. 4b). The active gel at 0h gave three broad lines of pseudoboehmite (Fig. 4c). On crystallization, the low angle peaks started to appear first followed by the other peaks. In the case of the low temperature batches 393K and 413K), after the disappearance of the 0h lines, mixtures of metavarsatite and varsatite appeared which then transformed into AlPO4-5. The intensity of the AlPO4-5 peaks increased with time. The maximum crystallinity of the sample was dependent on the reaction temperature.
The most
crystalline sample was obtained at 4.5h at 473K. The scanning electron micrographs show that the 0h flakes (Fig. 1c) dissolved in the template and formed small particles (Fig. 1d: after 2.25h). These aggregated later (Fig. 1e: after 3.5h). These aggregates with clear surface edges transformed into cauliflower shaped crystals (5m; Fig. 1f; after 4.5h). The apparent activation energy for nucleation, En was calculated from the expression [13], dln(1/n)/d(1/T) = - En/R, where 'n' is induction period, viz., the point on the crystallization at which conversion to the crystalline phase just occurs. Similarly, Ec, the apparent activation energy for crystal growth was calculated from the temperature dependence of the rate of crystallization and was obtained from the point in the 46
crystallization curve at which crystallization was 50%. represented as, dln(1/)/d(1/T) = - Ec/R, where
The rate equation can be
represents the time in hours for 50%
crystallization. The calculated value of the energy of nucleation (En) was 9.34 KJ/mole and the energy of crystallization (Ec) was 3.68 KJ/mole. The crystallization curves can be described mathematically
by the Avrami-
Erofeev equation [14,15] as ln[1/1-] = ktm where and t are functional conversion and reaction time respectively, while k and m are constants.
To check that the zeolite
formation in our case conforms to the above general picture, the data were fitted to the above equation and the results presented in Fig. 7 reveal good fit. The values of k and m obtained from the linear plots are also given in Fig. 7. The increase in the values of k and m with an increase in temperature indicate that the rates of nucleation and crystallization of AlPO4-5 increase with temperature. Fig. 6 shows that the rate of nucleation and crystallization are very high during AlPO4-5 synthesis using hexamethyleneimine. Under similar conditions, nearly 100% crystallization in the presence of triethylamine was reached only at 24h [16] for AlPO4-5 and 6h for SAPO-5 [17]. It is observed that at low temperatures (473K) and on prolonged crystallization, dense phases of AlPO4.xH2O and other crystalline impurities are formed along with AlPO4-5. The dense phases are tridymite, berilinite and cristobalite. It appears that at high reaction temperatures, the mobility of the templating organic species increases and that they come out of the pores of the framework structures easily. This may result in instability of the host structure as evidenced by the appearance of the dense phase impurities at high temperatures. The formation of aluminophosphate molecular sieves occurs from a gel to a layer to a three dimensional structure.
Here we have reported the synthesis of a layer
phosphate AlPO4-L and demonstrated that it is the intermediate phase in the formation of AlPO4-5, AlPO4-22, AlPO4-16 and SAPO-35. Such a layer structure formation can be schematically shown as in Fig. 3.8. The porous layer may undergo potential hydrolysis with interaction with solvent and form smaller layer chains. These in turn undergo interaction and self condensation to form different structure types.
47
Fig. 3.1: SEM photographs at different crystallization times, synthesis gel (Al2O3: P2O5: 1.16 HEM: 45 H2O) at a) 0 h, b) 2.5 h, c) 3.5 h, and d) 24 h; e) SAPO-5; f) AlPO4-5 (non-aqueous).
48
Fig. 3.2: X-ray diffraction patterns for AlPO4-5 in non-aqueous medium at a) 5th day, b) 10th day, c) 15th day.
49
Fig. 3.3: Variation in crystallinity of AlPO4-5 (non-aqueous) with duration of synthesis.
50
Fig. 3.4: X-ray diffraction patterns recorded at different time intervals: a) Catapal B, b) aluminophosphate gel ( C = crystobalite, B = berilinite and T = tridymite), c) synthesis gel (Al2O3: P2O5: 1.16 HEM: 45 H2O) at 0 h, d) 2.30 h, e) 4h, f) 8h, g) 16 h, h) 24 h and i) 36 h.
51
Fig. 3.5: Variation in pH during the synthesis at different temperatures a) 293 K, b) 313 K, c) 433 K, d) 453 K and e) 473 K. 52
Fig. 3.6: Variation in crystallinity with duration of synthesis at different temperatures. A) 293 K, b) 313 K, c) 433 K, d) 453 K and e) 473 K.
53
Fig. 3.7: Avrami Erofeev plots at different temperatures a) 393 K, b) 413 K,c) 433 K, d) 453 K and e) 473 K. Aavrami-Erofeev parameters are 393 K, k = 0.945, m = 32.83; 433 K = 1.510, m = -2.93; 493 K, k = 1.90, m = -1.94.
54
Fig. 3.8: Tentative routes for aluminophosphate formation through AlPO4-L.
55
Table 3.1: Influence of variation of template concentration on the nature of the crystalline product. Moles of template 1.16
1.35 2.32
Time of crystallization (hours) 4 48 120 192 24 24 148
Product
AlPO4-5 AlPO4-5 + AlPO4-L AlPO4-L + AlPO4-22 AlPO4-22 AlPO4-16 AlPO4-L AlPO4-L
AlPO4-L – a novel layered aluminophosphate
56
Table 3.2: Synthesis of AlPO4-L at different conditions S. No. 1.
2.
3.
4.
5.
6.
7.
Gel molar composition Al2O3: P2O5: 2.32R: 45H2O Al2O3: P2O5: 2.32R: 45H2O Al2O3: P2O5: 1.16R: 45H2O Al2O3: P2O5: 1.35R: 45H2O Al2O3: 1.8P2O5: 4.5R: 45EG 0.3SiO2: Al2O3: 1.8P2O5: 4.5R: 45EG Al2O3: P2O5: 2.32DPA: 40H2O
Al-source Catapal-B
Crystalliztion Phase Crystallization Phase time obtained time obtained 2d AlPO4-L 11d AlPO4-L
Al-iso propoxide
2d
AlPO4-L 11d
AlPO4-L
Catapal-B
5d
AlPO48d L+ AlPO422 AlPO4-L 24h AlPO416
AlPO422
Al5d isopropoxide
AlPO4-L 15d
AlPO4-5
Al5d isopropoxide
SAPO-L 15d
SAPO35
Catapal-B
AlPO4-L -
-
Al16h isopropoxide
2d
57
AlPO416
Table 3.3: Influence of variations in synthesis parameters on pH and crystalline phase. Parameter
Molar ratio
Initial pH
Final pH
(a) Al/P
0.8 0.9
3.54 3.71
6.88 7.55
1.0 1.1
4.46 3.72
6.92 9.34
1.2
4.19
9.17
30 50 75
4.39 4.46 3.87
7.17 6.74 7.30
100
3.63
7.14
125
3.48
6.85
0.0 0.5
4.46 3.22
6.92 6.80
1.0 1.5
2.12 1.58
5.84 1.89
(b) water/P2O5
© HCl/P2O5
C = cristobalite, T = tridymite and B = berilinite
58
Crystalline phase AlPO4-22 AlPO4-5 + AlPO4-22 AlPO4-5 AlPO4-5 + impurity AlPO4- + impurity AlPO4-16 AlPO4-5 AlPO4- 5 + AlPO4-22 AlPO4- 5 + AlPO4-22 AlPO4- 5 + AlPO4-22 AlPO4-5 AlPO4-5 + C,T,B C,T,B C,T,B
Table 3.4 : Synthesis of VAPO-5 and VAPSO-5 molecular sieve. S.No. 1. 2. 3. 4. 5. 6.
Gel composition Al2O3: P2O5: 1.16 HEM: 0.01-0.08 V2O5: 45 H2O Al2O3: P2O5: 1.16 HEM: 0.1 V2O5: 45 H2O Al2O3:0.99-0.94 P2O5: TPA: 0.01-0.06 V2O5: 40 H2O Al2O3: 0.92P2O5: TPA: 0.08 V2O5: 40 H2O Al2O3: P2O5: 1.16 HEM: 0.1 SiO2: 45 H2O Al2O3: P2O5: 1.16 HEM: 0.2-1.0 SiO2: 45 H2O
7. 8. 9.
Al2O3: P2O5: TPA: 0.1-1.0 SiO2: 45 H2O Al2O3: P2O5: 1.16 HEM: 0.08 V2O5: 0.1-1.0 SiO2: 45 H2O Al2O3: 0.96P2O5: TPA: 0.04 V2O5: 0.1-1.0 SiO2: 45 H2O
59
Product VAPO-5 C+T+B VAPO-5 C+T+B SAPO-5 SAPO-5 + impurity SAPO-5 VAPSO-5 VAPSO-5
60
61
REFERENCES
1.
Sykes, P., ―Mechanism in organic chemistry‖, Orient Longman, London, 71 (1981).
2.
Wilson, S.T., Lok, B.M. and Flanigen, E.M. U.S. Pat. 4,310,440 (1982).
3.
Bennet, J.M., and Kirchner, R.M. Zeolites, 11, 503 (1991).
4
Lohse, U., Vogt, F., and Richter-Mendau, J. Cryst. Res. Technol., 28, 1101 (1993).
5.
Lok, B.M., Messina, C.A., Patton, R.L., Gajek, R.T., Cannon, T.R., and Flanigen, E.M. U.S. Pat. 4,440,871 (1994).
6.
Richardson Jr., J.W., Pluth, J.J., and Smith, J.V., Naturwissenschaften, 76, 467 (1989).
7.
Cultaz, A., and Sand, L.B., Adv. Chem. Ser., 121, 140 (1973).
8.
Avarami, M.J., Chem. Phys. 9, 117 (1941).
9.
Erofeev, B.V., C.R. Acad. Sc. USSR, 52, 511 (1946).
10.
Weyda, H., and Lechert, H., Zeolites, 10, 251 (1990).
11.
Kamble, K.R., Ph.D., thesis on ―Synthesis., Characterization and Catalytic reactions over metalphosphate molecular sieves‖, Submitted to Poona University, Pune, India, p. 62 (1990).
62
CHAPTER IV ________________________________________________________________________ PHYSICOCHEMICAL CHARACTERIZATION ________________________________________________________________________
63
4.1: INTRODUCTION An understanding of the crystallization mechanism of aluminophosphate molecular sieves is expected to help in the synthesis of new molecular sieves. Although some precursor phases for specific structures have already been reported in the literature, the present studies have resulted in the isolation of a common precursor named as AlPO4L, in the synthesis of a number of structures (AlPO4-5, -16, -22 and -35) using hexamethyleneimine as the template. The novel material is characterized by XRD, SEM, FT-IR and TGA/DTA, which give the structural features of AlPO4-L. Encapsulated hexamethyleneimine present in the as-synthesized samples of AlPO4-5, -16, -22 and -35 was characterized by carbon and nitrogen analysis, TGA/DTA, TG-MS and
13
C NMR
spectroscopy. Also the incorporation of silicon in ethylene glycol medium has been attempted to improve silicon incorporation in a catalytically favorable mode, and to improve crystallinity. The incorporation of silicon in SAPO-35 has been studied by chemical analysis, XRD, SEM, FT-IR and
29
Si NMR techniques.
Further, the
incorporation and location of vanadium in VAPO-31, VAPO-41, VAPSO-5 and VAPO-5 has been examined by chemical analysis, XRD, SEM, FT-IR, ESR, UV-Vis, MANSMR and cyclic voltammetry. 4.2: ELEMENTALANALYSIS Elemental analysis of the different samples in the as-synthesized form namely, AlPO4-5, AlPO4-16, SAPO-35, VAPO-31, VAPO-5 and VAPSO-5 with different vanadium contents was carried out by chemical analysis, atomic absorption spectroscopy, X-ray fluorescence spectroscopy, inductively coupled plasma spectroscopy and microanalysis techniques. 4.2.1 : Aluminophosphate molecular sieves The elemental composition of AlPO4-5, AlPO4-16, AlPO4-22, AlPO4-31 and SAPO-35 are similar to the reported values1,2. Calcined AlPO4-5 on saturation with hexamethyleneimine sorbed 0.59 molecules per unit cell. The as-synthesized form of AlPO4-5, synthesized using hexamethyleneimine from aqueous medium contained two templates per unit cell. However, AlPO4-5 synthesized using the same template in nonaqueous medium contained three templates per unit cell. This is sterically possible if one template is packed tight over the other two along the c-direction of the unit cell. The template is less tightly packed in the case of the AlPO4-5 synthesized in aqueous medium as two template molecules are present per unit cell in the sample. In the case of AlPO 422, four hexamethyleneimine molecules were located inside a unit cell, probably two each per RPA cage, the AWW cage being very small, it is not likely to accommodate a template molecule. AlPO4-16 also contained four template per unit cell, each template probably being present in one cage of the structure. SAPO-35 contains four cages per unit cell,
each cage occluding one hexamethyleneimine molecule.
The novel
aluminophosphate, AlPO4-L possess a high template pore volume of 0.33 ml/g. 4.2.2: Vanadium aluminophosphate molecular sieves 4.2.2A: VAPO-5 and VAPSO-5 The V/V+Al+P ratios of VAPO-5 samples with different vanadium contents synthesized using HEM and TPA templates are listed in Tables 4.1 and 4.2. Elemental analysis of the ammonium acetate treated samples is presented in Table 4.2. The molar ratio of Al/V+P = ~1, indicates the replacement of phosphorous ions by vanadium (Table 2). Eventhough, the vanadium input values are the same for both the templates, the VAPO-5 samples synthesized using HEM contain more vanadium compared to those synthesized using TPA (Table 1).
Besides, the maximum amount of permissible
vanadium (to ensure crystallization) in the gel is different for the two templates, the maximum V/V+Al+P values for HEM and TPA being 0.0657 and 0.0339, respectively. The maximum amount of vanadium incorporation through synthesis was 0.2V/UC (unit cell) and 0.1V/UC for HEM and TPA respectively. Ammonium acetate washing reduces this to 0.05V/UC and 0.04V/UC (Table 4.2). Addition of excess vanadium leads 64
to dense phases like tridymite, cristobalite, and berilinite. About 60% of the vanadium present in the gel appears to be present in the VAPO-5 samples irrespective of the template. The % of vanadium picked from the gel decreases with increase in vanadium concentration in the gel (Table 4.2). On ammonium acetate treatment, about one third of the vanadium is retained in the solid. The leaching of vanadium is more in the VAPO-5 samples prepared using HEM. Vanadium and Si contents in VAPSO-5 samples are presented in Table 4.3. At low Si content, the vanadium content is higher in the VAPSO-5 samples than in the starting gel. At higher Si content, the V pickup decreases (Table 4.3). The presence of silicon did not affect the crystallization of the VAPSO-5, though a small reduction in XRD crystallinity of the samples was noticed. After ammonium acetate treatment, the vanadium content of the VAPSO samples are similar to those of VAPOs for a given V content in the synthesis gel at lower Si contents (Table 4.3). A decrease in V pick up is noticed when the Si/T (T=V+Si+Al+P) ratio is greater than about 0.12. Apparently, Si influences V-incorporation. The V4+ content estimated by the titration procedure in the as-synthesized and calcined VAPO-5 and VAPSO-5 samples is presented in Table 4.4. It is observed that both the as-synthesized and calcined samples contain large amounts of V4+ ions irrespective of the template used in the gel. However, on calcination, the V4+content decreases, the decrease being less in the VAPSO samples. The yields of the crystalline VAPO and VAPSO samples prepared using HEM and TPA are reported in Table 4.5. The yields were calculated based on the theoretical yields (based on input quantities) of VAPO-5 and VAPSO-5. The yields increase with increasing V content of the gel. 4.2.2B: VAPO-31 and VAPO-41 Just as in VAPO-5 samples, vanadium replaces phosphorous ions in VAPO-31 and VAPO-41 frameworks. Syntheses of VAPO-31 and VAPO-41 were carried out using typical diamine templates. An increase in vanadium loading in the gel increases the vanadium content in the crystalline phase. In VAPO-41, maximum V2O5 was incorporated in the framework when the gel composition was 0.02V2O5: Al2O3: P2O5. Further increase in vanadium led to VAPO-11, when diethylamine template is used. Batches with di-n-pentylamine produced a dense phase. CN – analysis reveals that VAPO-31 synthesized from both di-n-propylamine (n-DPA) and di-n-butylamine (nDBA) contains 1.5 template molecules per unit cell. AlPO4-31 has three channels per unit cell parallel to the c-direction with a repeat distance of 5Å. Our calculations ( using Insight II software supplied by MSI Inc. U.S.A)
show that the dimensions of the
template molecules are: n-DPA = 10.24 x 3.3 x 3.0 Å; n-DBA = 12.5 x 3.3 x 3.0 Å and nDpeA = 15.2 x 3.3 x 3.0 Å. It appears that these template molecules occupy two or more unitcell lengths of the straight channel. Though both n-DPA and n-DBA cause the crystallization of the AlPO4-31, n-DPeA does not. The reason may be that the length of the molecule has an effect on the crystallization of AlPO4-31. Chemical analysis of VAPO-31 synthesized from di-n-propylamine shows the composition: Al2O3: 0.99P2O5: 0.014V2O5. AlPO4-41 consists of four diethylamine template molecules per unit cell. VAPO-41 has an elemental composition: Al2O3: P2O5: 0.006V2O5. 4.3: X-RAY DIFFRACTION The X-ray diffraction patterns of AlPO4-5, AlPO4-16, AlPO4-22, AlPO4-31, AlPO4-L and SAPO-35 are presented in Fig. 4.1. The X-ray diffraction patterns of AlPO4-5, AlPO4-16, AlPO4-22, AlPO4-31, AlPO4-L and SAPO-35 match well with the published patterns [5,6,8]. The XRD pattern of AlPO4-L did not match any of the reported aluminophosphates and their analogs. The appearance of three intense peaks at low values of (6.4, 12.6 and 18.6o), suggests that it is a short range ordered lamellar type aluminophosphate phase [9]. Further, it was observed as a precursor phase during the synthesis of AlPO4-5, AlPO4-16, AlPO4-22, AlPO4-31 and SAPO-35. AlPO4-l is found to be thermally stable only up to 473K, though the silicon containing sample is 65
stable up to 573K. The h k l values of the synthesized aluminophosphates were reported elsewhere [10]. The X-ray diffraction patterns of VAPO-5, VAPSO-5, VAPO-31 and VAPO-41 samples are presented in Fig. 4.1. They match well with the reported ones [1-2]. It is found that no crystalline impurity is present in the synthesized phases. In the case of VAPO-5 samples synthesized using HEM, the crystallinity decreased on increasing the vanadium concentration; on the contrary the crystallinity increased while using the tripropylamine template (Table 4.5). However, the crystallinity of VAPSO-5 decreased with increase in silicon content in the case of both templates. 4.4: SCANNING ELECTRON MICROSCOPY The morphology and particle size of the aluminophosphates, AlPO4-5, AlPO4-22, AlPO4-16, SAPO-35, AlPO4-L, VAPO-31 and VAPO-41 are presented in Table 4.6.The SEM photographs (Fig. 4.4 and 4.5) show that the morphology and particle size depend on the type of material. Though at low concentrations of V (samples A and B), a decrease in particle size is noted compared to AlPO4-5, subsequent increase in V-content tends to increase the particle size of the crystallites (samples C & D). Additional silicon incorporation decreases the particle size (except for sample A) (Table 4.7). Vanadium and/ or silicon incorporation also changes the morphology of the AlPO4-5 samples (Fig. 4.5). 4.5: THERMAL ANALYSIS 4.5.1: Aluminophosphate molecular sieves 4.5.1A: AlPO4-L TG/DTA curves (Fig. 4.6) show the presence of three stages of weight loss. Table 1 presents a comparison of weight loss of AlPO4-L with AlPO4-5 prepared from the same template. The large endothermic weight loss (17% in the range 293-558K with an endothermic peak at 384K) in the first stage reveals the presence of more water molecules along with template.
The two exothermic peaks at 688 and 840K are
associated each with weight loss of 8%. Compared to AlPO4-5, the total weight loss is more (33% vs. 17.9%) for AlPO4-L, as the template content is more (1M vs 0.3M for Al2O3: P2O5). Besides the oxidative decomposition occuring at higher temperatures (688, 840K for AlPO4-L vs. 633, 747K for AlPO4-5 [10]) supports the presence of charge compensated templates in the former. 4.5.1B: AlPO4-5, AlPO4-16 and AlPO4-22 The TG/DTA plots are presented in Fig. 4.7. The weight loss of AlPO4-5, AlPO416, and AlPO4-22 occurs in three to five stages (Table 4.9). The first stage endothermic loss at 100oC is due to loss of physisorbed water and template. In the case of AlPO4-5 and AlPO4-22 the physisorbed material loss occur in a single step. However, AlPO4-16 loses the same in two different stages at 93 and 133oC due to its very small pore openings. The oxidative decomposition of hexamethyleneimine occurs in three stages in AlPO4-16 and AlPO4-22 as against two stages in AlPO4-5. AlPO4-5 is a large pore molecular sieve. AlPO4-16 and AlPO4-22 are small and very small pore molecular sieves. The small pores make the oxidative decomposition and elimination of products difficult and leads to an extra combustion stage. The higher temperature required for the elimination of the template in AlPO4-22 (381, 493, and 841oC vs 319, 470 and 837oC for AlPO4-16) is due to the presence of partially ionized templates. A partial amorphization due to the loss of chemisorbed water molecules occurs at about 800oC for AlPO4-22 and AlPO4-16 molecular sieves. Temperature programmed desorption mass spectra (TPD-MS) of the assynthesized AlPO4-5 and AlPO4-22 carried out in an inert atmosphere (argon) show that the template hexamethyleneimine is desorbed only after fragmentation. Mainly water and NH3 are evolved below 202oC. Water, NH3 and methane are the main products of the decomposition. The decomposition of the template (hexamethyleneimine) in inert atmosphere occurs below 600oC into small fragments like CH4, C2H4, C3H6 and NH3 66
(Fig. 4.4). The above fragments account for a loss of 11.5% and 6.5% (for AlPO 4-5 and AlPO4-22).
As these (especially NH3 and CH4) are richer in H2 than the parent
compound, a C-rich skeleton remains on the AlPO4s which desorbs at a much higher temperature. Mass spectral data also reveal that very little undecomposed template is desorbed. The relatively higher temperature of desorption of fragmentation products for AlPO4-22 compared to AlPO4-5 (502oC vs 452oC) is attributed to the presence of some charge compensating in the AWW structure. TG data shows that AlPO4-22 continuously loses H2O beyond 702oC due to dehydration and amorphization unlike AlPO4-5, which is thermally more stable. 4.5.1C: SAPO-35 The TG/DTA curves of the template containing SAPO-35 sample (synthesized in ethylene glycol) are presented in Fig. 4.3e. In the low temperature region up to 529 K, there is an endothermic weight loss of 1.6%, mainly due to loss of adsorbed water. In the temperature region 256 to 860oC,
there are three stages of exothermic weight loss
(0.8%, 7.8%, and 8.4%) due to the oxidative decomposition of the organic template occluded in the pores with probably a part of the hydrogen burning. In the region 415630oC, protonated template molecules interact strongly with the framework charge. The fragments of templates, decomposed and adsorbed at lower temperatures, are partially oxidized leaving behind a coke residue. In the last stage (630-860oC), this coke is eliminated by oxidation. Carbon and nitrogen analysis (13.50%C and 2.63%N) shows that no ethylene glycol molecule is involved in the structure build up. 4.5.2: Vanadium aluminophosphate molecular sieves 4.5.2A : VAPO-5 and VAPSO-5 The TG/DTA in air curves of as-synthesized AlPO4-5, VAPO-5 and VAPSO-5 synthesized using HEM are presented in Fig. 4.4. The TGA results are also presented in Table 7. The oxidative decomposition of the template occurs in multiple stages. AlPO45 looses the entire template by oxidative decomposition in two stages at 360 and 474oC. VAPO-5 and VAPSO-5 lose the occluded templates in three stages. If the vanadium ions (V4+ present in the as-synthesized VAPO-5 and VAPSO-5 samples) replace P5+ ions in the structure, an extra negative charge will be created for every substitution. The charge will be then compensated by the template molecules in the RNH2+ form. In such a case, the decomposition of the charge compensating template can be expected to occur at higher temperatures and probably the third stage combustion corresponds to the removal of these molecules. In the case of VAPSO-5, the replacement of P5+ by Si4+ creates additional negative charges, which can also be charge compensated by the template ions. As a result the template content is higher in VAPSO-5; besides, a fourth combustion stage is also found. 4.5.2B: VAPO-31 and VAPO-41 The TG/DTA patterns of the as-synthesized samples of AlPO4-31, V2O5/AlPO431, VAPO-31, AlPO4-41 and VAPO-41 are presented in Fig. 4.11. Aluminophosphates lose weight in three stages, V2O5 supported on aluminophosphate exhibits only two stages of weight loss and VAPO molecular sieves lose weight in four stages (Table 4.12), revealing the consequence of vanadium incorporation in the framework of aluminophosphates. In all the molecular sieves, the first stage endothermic weight loss at ~373K is due to the loss of physisorbed water and template molecules. The occluded templates are lost exothermally by oxidative decomposition in two stages at 592 and 647K (for AlPO4-41) and at 614 and
636K (for AlPO4-31).
However, the V2O5
supported molecular sieve eliminates the template in a single step with a peak maximum at 629K. The vanadium-incorporated AlPO4s eliminate the template in three stages of oxidative decomposition at 599, 643, and 714K (VAPO-41), and 569, 588 and 662K (VAPO-31). The three stage decomposition could indicate the absence of amorphous or agglomerated vanadium in the as-synthesized VAPOs (as the V2O5 impregnated sample had a single stage). The third exothermic peak at higher temperatures is probably due to 67
the charge compensating template (RNH2+) associated with V4+ ions replacing P5+ ions. As the ionic bonded RNH2+ species is more stable, it is difficult to decompose and is eliminated at higher temperatures. 4.6: INFRARED SPECTROSCOPY 4.6.1. Framework vibrations IR spectra recorded in the framework region of the aluminophosphates, AlPO4-5, AlPO4-16, AlPO4-22, AlPO4-31, AlPO4-L and SAPO-35 are presented in Fig. 4.12. Although, the tetrahedral pore opening vibrations around 480 cm-1 are not sharp in AlPO4-22, AlPO4-16 and AlPO4-L, the T-O-T bending vibrations (T = Al or P) at 532 cm-1 are clear.
The T-O-T symmetric stretching vibrations around 1125, 1173(sh),
1240(sh) cm-1 (Table 4.13) are also observed. The double 4,6 ring vibrations at about 655, 710 and 731 cm-1 are clear in SAPO-35 and AlPO4-L. Although the same vibrations are also present in AlPO4-16, the two peaks are merged in AlPO4-22 revealing a single peak in this region. 4.6.2: Hydroxyl region The transmittance spectrum of the calcined sample of SAPO-35 (synthesized from ethylene glycol) in the hydroxyl stretching region is presented in Fig. 4.13. In order to understand the nature of these surface hydroxyl groups, studies reported earlier on some small pore molecular sieves may be considered. The spectra of SAPO-17, 18 and 34 [6] consist of isolated Al-OH (3769, 3793 cm-1), Si-OH (3740, 3720 cm-1) and P-OH (3764 cm-1) absorption bands , which are of much lower intensities compared to the bridging hydroxyl groups, Si-(OH)-Al. Whereas, HLZ-132[7] shows strong bands at 3740 and 3720 cm-1 due to external and internal Si-OH groups, in the case of the isostructural SAPO-35, these isolated hydroxyl groups are almost negligible. There are two bridging hydroxyl groups in SAPO-17 (3600, 3587 cm-1) and SAPO-34 (3624, 3601 cm-1) located in two different positions, the higher frequency band in spacious cages and the lower frequency band in smaller cages. In the case of HLZ-132 [7] and SAPO-35, there are three bridging hydroxyl groups that in SAPO-35 being better resolved and more intense. In another isostructural molecular sieve DAF-4. Barrett et al. [8] have reported only one kind of bridging hydroxyl group at 3580 cm-1. The bridging hydroxyl groups due to Si incorporation in AlPO4 ,molecular sieves can arise from different locations.
The
Bronsted acid, bridging hydroxyl groups are generated only when the replacement of P by Si occurs.
31
P MASNMR studies indicate two types of P environments, one in single
6 membered ring (S6R) and the other in double six membered rings (D6R) [12]. Therefore, when Si replaces these two P atoms in different chemical environments, the bridging hydroxyl groups arising there from should also have two different stretching vibration frequencies. OH groups in S6R have absorption bands at higher wave numbers than those in D6R positions [6]. But again, among the OH groups belonging to D6R, those in bigger cages but nearer to S6R absorb at higher wave numbers than those present within the D6R itself [6]. In this way, three types of bridging hydroxyl groups can be expected to be present in SAPO-35 molecular sieves. Therefore, the bridging OH groups with stretching vibrations at 3631, 3604 and 3580 cm-1 can be assigned to OH groups in bigger cages near S6R, in bigger cages near D6R and those actually confined inside the D6R, respectively. The FT-IR spectra of chemisorbed NH3 on SAPO-35 in the temperature range of 323-673 K are presented in Fig. 4.14. Chemisorption of NH3 leads to NH4+ on Bronsted acid sites and co-ordinatively bound molecules on Lewis acid sites. NH3 chemisorbed on Bronsted acid sites in SAPO-35 gives two pairs of bands in N-H symmetric stretching vibrations, besides a 3-fold degenerate band centered at 1460 cm-1 due to the deformation vibrations of the N-H bond in the NH4+ species. The two pairs are observed at 3380, 3280 and 2925, 2746 cm-1. In comparison with zeolites [9-10] and SAPO-40 [11], the pair of bands at 3380 and 3280 cm-1 is assigned to stretching vibrations of free N-H bonds and the second pair at 2925 and 2720 cm-1 to N-H stretching modes involved in H68
bonding. The shift to lower frequency (455 and 460 cm-1) due to H-bonding in SAPO35 is on the whole larger than that observed in the case of H-Y and H-mordenite (and 470 cm-1) [9,10]due to the higherbasicity of the framework O atoms in SAPO-35.
A
comparison of the shift of the bands in the case of SAPO-40 (735 and 820 cm-1) [11] with the shift observed in SAPO-35 suggests that the latter is less basic (or more acidic) than SAPO-40. In the region of deformation vibrations, we donot see any and due to coordinatively bound NH3 and bands at 1500 (sh), 1455(broad) and 1400 (sh) cm-1 which are due to Bronsted acid sites are observed, as has been reported for SAPO-40 [11]. The position of this composite peak which is observed at 1455 cm-1 in the spectrum at 323 K is shifted to 1445 cm-1 at 573 K due to a temperature effect. The absence of Lewis acid sites and silanol groups on the external surface and defect sites suggests the absence of silica microdomains on the surface of SAPO-35, which would have arisen had there been considerable substitution of a pair of Si atoms for a pair of P and Al atoms in the framework. Therefore FTIR studies strongly indicate that Si atoms mainly substitute P atoms in the SAPO-35 sample synthesized using ethylene glycol. 4.7: NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY 4.7.1: 13C NMR The 13C NMR spectrum of pure liquid hexamethyleneimine (Fig. 4.15a) consist of three sharp lines at = 27.60 and 49.74 ppm due to the three non equivalent pairs of catoms viz., C3/C4, C2/C5, C1/C6 respectively.
In the case of hexamethyleneimine
occluded in AlPO4-16, AlPO4-5 , AlPO4-22 , the two lines (Fig. 4.15) in the high field region are not resolved resulting in a broad single double intense line at ~39.5ppm and a single low intensity line 60.00ppm due to restriction in the movement of the template inside the molecular sieves. The channel dimension of AlPO4-5 (cross section 7.3Å and c-repeat distance 8.4Å) can easily accommodate the hexamethyleneimine molecule, where as in AlPO4-22 which has one PO3(OH)2- species in each AWW cage (or for any two RPA cages) protonated hexamethyleneimine should be assigned a unique position in each of the adjacent RPA cages, thus electronically satisfying the excessive –ve charge in the AWW cage.
Such charge compensation probably plays an important role in
stabilizing the AlPO4-22 structure enclosing such a tight fitting template. In addition, it should be noted that the dimension of the RPA cage is just sufficient to fit the hexamethyleneimine molecule. 4.7.2: 27Al and 29Si NMR The
27
Al and
29
Si MAS NMR spectra of SAPO-35 are presented in Fig. 4.16.
The spectra for the as-synthesized and calcined samples (Fig. 4.16a and b) show a single symmetrical 27Al line at delta = 36.26 ppm with respect to Al(H2O)63+ indicating a single tetrahedral aluminium species confirming the high purity of the sample and revealing the high thermal stability of Al cations in the SAPO-35 framework. A similar single line spectrum of
27
Al (at 29.7 ppm) was also reported by Blackwell et al. [12] for the as-
synthesized SAPO-35 sample (synthesized in aqueous medium). In the same study, they have reported a single peak at delta = -91.5 ppm for 29Si, which was attributed to Si(4Al), whereas the sample of the present study exhibits two peaks (Fig. 4.16) at -89.9 and -95.15 ppm. The two peaks are probably due to the substitution of P present in two different locations in double 6 ring (D6R) and in single 6 ring (S6R), respectively, in the Levynite like structure of SAPO-35. The above assignment is based on earlier reports that two lines are observed for the
31
P spectrum in SAPO-35 assigned to P atoms in these two
positions [12]. However in the case of the calcined sample (Fig. 4.16), the peaks become broad overlapping each other. The 27Al and 29Si MASNMR spectra of VAPSO-5A and VAPSO-5B synthesized from hexamethyleneimine is presented in Fig. 4.17.
27
Al NMR spectra of VAPSO-5
samples show a dominant line at = ~ 42 ppm ascribed to tetrahedrally co-ordinated aluminium atoms bonded via four phosphorous atoms [13]. The value of the chemical shift is in the range observed for other silicoaluminophosphates [12,14-16]. The small 69
line at ca. 3.0 ppm is most probably caused by octahedrally co-ordinated aluminium atoms of unreacted starting material [17]. The additional line at = ~ -11.5 ppm is probably due to aluminium atoms bonded to four phosphorous atoms (via oxygen), but possessing a higher co-ordination due to interaction with adsorbed water molecules. This can be eliminated on thermal dehydration [12, 15-16].
29
Si MASNMR shows a
prominent line at –111.23 and –111.78 ppm due to silicon atoms surrounded by four silicon atoms [18-20]. However, the presence of additional unresolved bands in the region 100-110 ppm reveals the presence of considerable amount of Si:2M:2Si and Si:M:3Si (where M = V or Al) species. 4.8: ELECTRON SPIN RESONANCE SPECTROSCOPY The electron spin resonance spectra of all the vanadium aluminophosphates were similar and their symmetry corresponds to VO2+ in a distorted octahedral or a square pyramidal environment [21]. On calcination, most of the vanadium is oxidized into V5+ ions. As a typical example, the ESR pattern of VAPO-5 (hexamethyleneimine) in the assynthesized, calcined forms is presented in Fig. 4.18. On calcination, the eight line hyperfine splitting almost disappeared and a single broad line spectrum was noticed. This is possibly due to some agglomerated form of vanadium species. Thermal reduction with hydrogen gave back the ESR spectrum similar to that of the as-synthesized sample, but the intensity was one sixth of that of the as-synthesized form. Refluxing in an organic solvent such as toluene or xylene also resulted in the reduction of the V5+ species. Treatment with NH4Oac solution (1M) at 393 K and calcination at 673 K after drying at 383 K lead to a spectrum with an intensity of ~5% of that of the synthesized sample. At liquid nitrogen temperature, the intensity of the spectrum was four fold of that obtained at room temperature suggesting that three fourths of the vanadium are present in a regular tetrahedral environment as such species are not expected to be seen at room temperature [22]. The intensity of the ESR lines increased with vanadium concentration for VAPO-5 and VAPSO-5.
Calcination of VAPSO-5 did not oxidize much of the V4+ ions as is
observed from the presence of hyperfine splittings with minor changes in intensity. A similar observation was also made in permanganometric titration (Table 4.13). The ESR parameters for the different samples are presented in Table 4.13. 4.9 UV-VIS SPECTROSCOPY Tetrahedrally coordinated V5+ ions exhibit a set of overlapping peaks at 239 and 343 nm. Octahedrally coordinated V4+ions have been reported to absorb at 443 nm. The as-synthesized VAPO-5 and SAPO-5 exhibit absorption bands at 288 nm due to Me-Olinkages. As-synthesized VAPO-5 samples exhibit three absorption bands at 280, 416 and 738nm corresponding to tetrahedrally coordinated V5+ ions [24,25] and octahedrally coordinated V4+ ions (d-d transitions ) [23,26]. The octahedrally coordinated V4+ ions disappear at lower vanadium concentrations. Calcined samples exhibit two peaks at around 280 and 415 nm, attributable to V5+ in tetrahedral and square pyramidal environments.
Square pyramidal vanadium is absent in samples containing less
vanadium. As-synthesized VAPSO samples give two absorption bands for tetrahedrally coordinated V5+ and distorted tetrahedral V5+ions. There is a small shift in the absorption frequencies on increasing the silicon concentration. These absorption bands are also present in the calcined samples, except that there is a small shift towards lower frequencies.
Similar to VAPO-5, VAPO-41 also possesses V5+ in tetrahedral and
distorted tetrahedral co-ordination and V4+ in octahedral environments. But in VAPO-31 , all the V5+ are present in a tetrahedral environment, a negligible amount of V5+ in distorted tetrahedral environment being present [Fig. 4.21]. The electron spin resonance spectra of all the vanadium aluminophosphates were similar and their symmetry corresponds to VO2+ in a distorted octahedral or a square pyramidal environment [12]. On calcination, most of the vanadium is oxidized into V5+ ions. As a typical example, the ESR pattern of VAPO-5 (hexamethyleneimine) in the as70
synthesized, calcined forms is presented in Fig. 4.5.
On calcination, the eightline
hyperfine splitting almost disappeared and a single broad line spectrum was noticed. This is possibly due to some agglomerated form of vanadium species. Thermal reduction with hydrogen gave back the ESR spectrum similar to that of the as-synthesized sample, but the intensity was one sixth of that of the as-synthesized form. Refluxing in an organic solvent such as toluene or xylene also resulted in the reduction of the V5+ species. Treatment with NH4Oac solution (1M) at 393 K and calcination at 673 K after drying at 383 K lead to a spectrum with an intensity of ~5% of that of the synthesized sample. At liquid nitrogen temperature, the intensity of the spectrum was four fold of that obtained at room temperature suggesting that three fourths of the vanadium are present in a regular tetrahedral environment as such species are not expected to be seen
at room
temperature24. The intensity of the ESR lines increased with vanadium concentration for VAPO-5 and VAPSO-5. Calcination of VAPSO-5 did not oxidize much of the V4+ ions as is observed from the presence of hyperfine splittings with minor changes in intensity. A similar observation was also made in permanganometric titration (Table 4.4). The ESR parameters for the different samples are presented in Table 4.8. Tetrahedrally coordinated V5+ ions exhibit a set of overlapping peaks at 239 and 343 nm. Octahedrally coordinated V4+ions have been reported to absorb at 443 nm. The as-synthesized VAPO-5 and SAPO-5 exhibit absorption bands at 288 nm due to Me-Olinkages. As-synthesized VAPO-5 samples exihibit three absorption bands at 280, 416 and 738nm corresponding to tetrahedrally coordinated V5+ ions25,26 and octahedrally coordinated V4+ ions (d-d transitions )13,27.
The octahedrally coordinated V4+ ions
disappear at lower vanadium concentrations. Calcined samples exhibit two peaks at around 280 and 415 nm, attributable to V5+ in tetrahedral and square pyramidal environments.
Square pyramidal vanadium is absent in samples containing less
vanadium. As-synthesized VAPSO samples give two absorption bands for tetrahedrally coordinated V5+ and distorted tetrahedral V5+ions. There is a small shift in the absorption frequencies on increasing the silicon concentration. These absorption bands are also present in the calcined samples, except that there is a small shift towards lower frequencies.
Similar to VAPO-5, VAPO-41 also possesses V5+ in tetrahedral and
distorted tetrahedral co-ordination and V4+ in octahedral environments. But in VAPO-31, all the V5+ are present in a tetrahedral environment, a negligible amount of V5+ in distorted tetrahedral environment being present [Fig. 4.21]. 4.10: CYCLIC VOLTAMMETRY Castro Martins et al. [42,43] have used cyclic voltammetric technique to prove the presence of Ti4+ ions in the framework of TS-1 and TS-2. Studies on the application of Cyclic voltammetry to V-molecular sieves are reported in this section. Besides VAPO molecular sieves, vanadosilicates (VS-1, and VS-2) and vanadium aluminosilicates (VAl-beta, V-Al-EU-1 and V-Al-ZSM-22) have also been used for comparison purposes. The above molecular sieves were obtained from co-workers in the laboratory and their synthesis and characterization have already have been reported [37-40, 44]. 4.10.1: VAPO-5, VS-1, VS-2 and V-Al-beta Typical cyclic voltammograms obtained for various vanadium containing molecular sieves coated on a platinum UME in 0.1MKCl (aqueous solution) at a representative scan rate of 50 mV/s and after 8 h of immersion are shown in Fig. 4.22. These steady-state voltammograms could be obtained only after ca. 8h of immersion of the electrode. This is presumably due to the complex evolution of response with time associated with diffusion in the spatially restricted molecular sieve system. Voltammogram of all molecular sieves soon after immersion resemble Fig. 4.22a (for silicalite and AlPO4-5) with no discernible redox peaks, and within 1-2h redox peaks originate with a slow increase of the peak current. Maximum peak currents are obtained after about 5-8 h of immersion, and final voltammograms (Fig. 4.22) are stable and 71
completely reproducible. The specific time required for observing steady state response is a function of porosity (5 h in the case of VAPO-5 with larger pores (7 Ao) and 8 h in the case of VS-1 with smaller pores (5.5 Ao) and this together with the increase of peak intensities with time shows that more and more vanadium sites are electro active owing to the progressive diffusion of ions through the microporous network.
Although the
voltammograms generally exhibit the presence of two reversible redox couples (except V-Al-beta and VAPO-31) with varying relative peak potentials (depending on the molecular sieve), the relevant current intensities are very different from one molecular sieve to another (Table 4.14). For example, in the case of VAPO-5C (Fig. 4.22e), the current intensities are about 2uA for a V/Al+P+V ratio of 0.010. In the case of VS-1 (Fig. 4.22b), the current intensities are about 100 uA for V/Si+V rtio of 0.011. In addition, the particle size of VS-1 is smaller than that of VAPO-5C. The non-Faradaic and ohmic contributions in these voltammograms are also different and it is not easy to relate the total amount of vanadium in each sample to the observed current intensities. The E1/2 values (described as the average
peak value of the oxidation and
reduction potentials) of the reversible redox peaks and peak potentials for the various vanadium containing molecular sieves are presented in Table 4.15. The two anodic peaks, the first peak at +0.38 (VS-1), +0.44 (VS-2), +0.33(V-Al-beta), and 0.34 (VPO-5) and the second peak at +0.16 (VS-1), +0.20 (VS-2) and +0.14(VAPO-5) indicate similarities in the anodic behavior for all the V-containing molecular sieves. However, in the case of the cathodic peaks although the first peak of the vanadium silicates +0.025 (VS-1), -0.025 (VS-2) and -0.038 (V-Al-beta) occurs at nearly the same potential, it is shifted by 0.3V in the case of VAPO-5 (+0.25) indicating that the vanadium species in these molecular sieves were easy to reduce in VAPO-5 compared with VS-1, VS-2 and V-Al-beta. The second cathodic peak shifted to a lower potential in the order VAPO-5 (0.04), VS-1 (-0.21) and VS-2 (-0.54 V). More over, AlPO4-5, silicalite, V2O5 and V2O5 supported molecular sieves (Fig. 4.23) are found to be electrochemically inactive in this potential region.
This implies that vanadium which has been introduced into the
molecular sieves during the synthesis and presumably incorporated into the framework could probably easily undergo facile redox transformation.
Further, the ammonium
acetate treated samples of VAPO-5C show a cyclic voltammogram similar to that of the calcined sample, suggesting a common source of redox behavior. In sharp contrast the V2O5 impregnated samples calcined at various temperatures (upto 1172 K) do not give any response in the working potential region (Fig. 4.23d).
In comparison, the as-
synthesized molecular sieves were inactive in this region, when UME was used. However, when the microelectrode (ME) was used at scan rate of 20 mV/sec a weak behavior similar to calcined samples was noticed. This indicates the presence of small amounts of electro active vanadium at the surface, the ones in the pores being blocked by the templates. In addition to the two redox peaks, an additional anodic peak is observed at +0.6V during the first few cycles, which subsequently vanishes (Fig. 4.24) in the case of VS-1 and VAPO-5. This signal is probably due to the leaching out of VO2+ ions (present as charge compensating ions in the channels of these molecular sieves) in exchange positions by K+ ions, possibly following the reaction [16]. VO2+ + H2O --------- VO2+ + 2H+ + e- (Eo = +0.77 V vs SCE) The standard electrode potential of 0.77V for the above reaction matches well the observed potential and this confirms the presence of cation exchange sites and VO2+ in the channels of VS-1, as has already been established previously.
However, AAS
analysis of the electrolytic solution (performed after the experiment) could not detect the presence of the leached out vanadium. The ammonium acetate treated samples also give an anodic peak at +0.6 V. This indicates that the VO2+ ions present in the channels are not removed by ammonium acetate treatment at room temperature. The non removal of VO2+ ions by NH4+ ions is not surprising, as generally polyvalent ions present in 72
exchange positions of zeolites are not easily exchanged by monovalent ions such as NH4+. The voltammograms taken in aqueous KCl solution are not discernible at high scan rates (higher than 1000 mV s-1). This is mainly due to the large IR drop and the increased contribution of non-Faradaic current making the background very broad at high scan rates. Interestingly, when tetra butyl ammonium tetrafluoborate (TBABF4) in acetonitrile is used as the electrolyte, a quasi-reversible behavior is observed [Fig. 4.25] which disappears after a few cycles. This is probably due to the leaching out of the surface bound extraframework vanadium into the electrolyte. If one assumed that the absence of two reversible peaks in this case is due to pore size restrictions preventing entry of the bulky TBA+BF4- ions into the framework, LiClO4 in acetonitrile should reveal the presence of two redox couples as observed in aqueous voltammograms. However, the use of LiClO4 instead of TBABF4 as a supporting electrolyte shows no marked changes in the voltammograms and the disappearance of the quasi-reversible peak after a few cycles confirms that the above assignment is due to the non-bonded extra framework vanadium. The reason for the lack of observance of the two signals in the case of acetonitrile is not clear. Table 4.15 presents information on the vanadium contents of the samples and the peak currents observed in the cyclic voltammograms. Although these two parameters cannot be directly correlated because of the difference in structure, particle size, interfacial area and ohmic drop, relative contributions of relevant redox species within one structure can be inferred by a careful analysis of this table. Both V4+ and V5+ ions are present in the as-synthesized VAPO-5 [28] and KVS-5[30] (equivalent to VS-1) while only V5+ ions present in the as-synthesized V-Al-beta. The E1/2 value (0.17) of the set of peaks observed for V-Al-beta is similar to that of the first set of peaks of VS-1, VS2 and VAPO-5 (E1/2 values being respectively, 0.20, 0.21 and 0.32V). This suggests that the first set of peaks is due to the V5+ ions are present in the as-synthesized samples and presumably these ions are present in the framework (as has been shown in the case of VAl-beta [37]). Additionally in the case of VAPO-5, the intensities of both signals 1 and 2 increase with increasing vanadium content (Fig. 4.26). However, it is noticed that the particle size of the samples decreases with increasing vanadium content and this may indirectly affect the variation in the peak currents. The standard electrode potential for the V5+/V4+ redox couple at pH 3-3.5 is +0.48V vs SCE [29]. On the basis of this and using available data [30] on the cyclic voltammograms of equimolar solutions (5 x 10-3 M; pH 4) of ammonium metavanadate and vanadyl sulfate which gives peaks at E1/2 = +0.07 and -0.14V (V5+/V4+ and V4+/V3+ redox couples, respectively), the voltammograms of the molecular sieves (except V-Albeta) can be explained by similarities of the molecular sieves. The first couple can readily be attributed as discussed earlier, to the V5+/V4+ redox cycle. However, the second couple could be due to either V4+/V3+ or environmentally different V5+/V4+. The latter possibility is indicated by the following experimental observations: (1) The V-Albeta sample which contains only V 5+ in the as-synthesized form [30], gives only the first set of redox signals. The other V-Containing molecular sieves give two sets of redox signals and are found to possess both V5+ and V4+ in the as-synthesized samples [17, 30] (Table 4.17). On calcination, the V4+ ions present are transformed into V5+ ions, and it is possible that the two sets of redox peaks arise from (i) the V5+ ions originally present in the as-synthesized samples and ii) the V5+ ions formed by the oxidation of the V4+ ions present in the as-synthesized samples. It is not unreasonable to believe that the V5+ ions (in the calcined samples) arising from two different sources have different environments and hence the different redox potentials. (2) The large size of the V3+ ions [radii of ions [32]: V3+(coordination number 6) = 0.64; V4+(5) = 0.53; V5+ (4) = 0.36; Si4+(4) = 0.26; Al3+ (4) = 0.39 and P5+ (4) = 0.17 Ao] and their preference for six co-ordination precludes their formation in the molecular sieve framework composed of smaller ions such as Si4+, 73
Al3+ and P5+. (3) Earlier workers have reported [33] two different redox peaks (shifted by ca 0.77V) for a environmentally different Co3+/Co2+ couple in the case of zeolite encapsulated Co-Salen complexes. Although this analogy is not strictly correct because the encapsulated Co-Salen complexes are in the extra framework region, the results clearly indicate the variation in the potential for the same couple with a change in environment. It, therefore, appears reasonable that the two sets of redox signals arise from two environmentally different V5+ species.
(4) Cyclic voltammograms of an
equimolar solution (5 x 10-3 M) of ammonium meta vanadate (V5+) and vanadyl sulfate (V4+) at pH 5.2 give E1/2values of 0.093 and -0.18 V (vs SCE), respectively for the V5+/V4+ and V4+/V3+ redox couples. The E1/2 values corresponding to V5+/V4+give an anodic shift of 45 mV for VAPO-5 and 105 mV for VS-1 if we assume that the second redox couple exhibited by the three molecular sieves corresponding to a V 5+/V4+ couple. The shifts in potential by considering a V4+/V3+ couple are much larger (VAPO-5 shows 225 mV shift for anodic signal and VS-1 shows 155 mV). An alternate explanation could be that the second set of redox peaks is due to the V4+/V3+ couple and originate from ions anchored to the framework surface. Studies on the as-synthesized VS-2 have revealed that V5+ ions are more strongly bound than V4+ ions and are mainly present in framework positions while a large number of V4+ ions are loosely bound to the framework [31]. In such a situation, the formation of bulky V3+ ions from the loosely bound vanadium species is possible as there may not be any steric constraints as these sites. As-synthesized V-Al-beta has been shown to contain only V5+ ions (all in the framework)and gives only one set of peaks (V5+/V4+). Thus it can be argued that the second couple arises from the V4+/V3+ transformation originating from the loosely bound ions (V4+ ions in the as-synthesized samples). Although such a possibility exists, this is less likely because one would expect the intensities of the second signal to be less than that of the first signal as the latter would have resulted from the V5+/V4+ transformation of all the V ions in the calcined samples while the former would have resulted from only the weakly bound (anchored) species. This is in contradiction with the experimental data given in Table 4.14. In order to understand the origin of the reversible behavior in the molecular sieves, two compounds having vanadium in known environments namely, vanadyl acetyl acetonate (square pyramidal) and vanadyl tripropoxide (distorted tetrahedral) were used as model compounds. Vanadyl acetylacetonate (Fig. 4.27(a) gives a reversible peak at E1/2
= -0.19 V together with an oxidation peak at VO2+(+0.80V), while vanadyl
isopropoxide (Fig. 4.27(b) gives a single reversible peak at E1/2 =+0.39V. The E1/2 value of vanadyl isopropoxide compares well with those of the first few V-containing molecular sieves studied (E1/2 = 0.17-0.32V). Furthermore, the presence of tetrahedral species (corresponding to the first redox couple) has been confirmed by EPR studies of reduced V-Al-beta [37] (Table 4.15). The redox cycle (E1/2 = -0.19 V) that appeared in the vanadyl acetylacetonate carbon paste modified electrode is closer to the V4+/V3+ redox couple (-0.18V) in solution phase at the same pH. It is difficult to compare this value with the second redox couple (E1/2 = 0.045 to -0.17 V) of the three molecular sieves (VS-1, VS-2 and VAPO-5). The EPR analysis shows the presence of V4+ in a square pyramidal symmetry in the as-synthesized samples of the above (VS-1, VS-2 and VAPO5) molecular sieves [26-28]. This analogy shows the possibility of the presence of V5+ in square pyramidal state in the above calcined molecular sieves. A comparison of the results of XRD, framework IR and EPR before and after separate coulometric reduction of VS-1 at -0.25 V and -0.0125V (vs SCE) illustrate several important features. For example, XRD patterns of the samples before and after coulomtric reduction showed no changes due to reduction of the V5+. This clearly indicates that the crystal structure is stable during the redox transformation. The amounts of charge passed during the coulometric reduction of VS-1 at constant potentials of 0.0125 V and -0.25 V are 11.45 and 47.68 C respectively. These values are very large 74
compared to the theoretical amount required for the total conversion of vanadium in the sample (5.79 C) indicating poor Faradaic efficiency if one assumes that all the vanadium sites are electrochemically accessible. Also the framework IR of the sample before and after coulometry does not reveal any significant changes in the spectra.
More
specifically, the intensity of the 960 cm-1 attributed to Si-O-V linkages in the framework remains the same before and after reduction. EPR spectra taken at 298 and 77K of VS-1, silicalite and graphite before and after coulometric reduction give further evidence for the above hypothesis of the presence of environmentally different V5+/V4+ redox couples. The coulometrically reduced VS-1graphite samples give a single peak (g = 2.000) and a hyperfine spectrum (Fig. 4.28(b); g1=1.993 and g2= 2.007) at 77 K. The EPR parameters of the above signals suggest the presence of [O2]- species [34-36], in agreement with the fact that adsorbed oxygen radical ions are formed by the reaction of V4+ (tetrahedral) ions with O2 as shown below. V4+(tetrahedral) + O2 --------- V5+.O2The intensity of the above signal increases with the potential (Fig. 4.28 (c)), revealing that these signals are not due to any impurities but are due to the reduction of V5+ to V4+. Further more, a comparison of the low temperature (77K) EPR spectra of electrochemically reduced samples of VS-1 and VS-1/graphite with those obtained after thermal reduction in a hydrogen atmosphere at 673K shows certain interesting features. For example VS-1 (after reduction in hydrogen) gives an eight line spectrum, while surprisingly such a spectrum is absent for similar samples with mixtures of VS-1 and graphite irrespective of whether the mixing was done before or after thermal reduction. In addition dilute samples of VOSO4 graphite and VOSO4 –graphite (calcined and H2 reduced) donot show any signal at 298 and 77K respectively despite the eight line hyperfine signal for all VOSO4 samples. It appears that the vanadium hyperfine splitting is suppressed by the electron current in graphite. Nevertheless, the EPR results confirm unambiguously the presence of V4+ after electrochemical reduction irrespective of the potential used for coulometry. From the above experiments, a probable route for the electron transfer reactions is as follows. V5+(OT)4 + K+ + e- 2C>1C. The results of the oxidation of nhexane over VAPO-31 and VAPO-41 are presented inTable 5.4. The secondary products viz. 2 and 3 hexanones are present in larger amounts revealing the excellent oxidation activity of vanadiumaluminophosphates. VAPO-31 is found to be a more active catalyst than VAPO-41 and is probably due to the higher incorporation of vanadium in the framework of VAPO-31 sieves. 5.5: BENZALDEHYDE ACETALIZATION Protection of the aldehyde group in benzaldehyde is an important reaction in organic synthesis during ring substitution reactons. Currently acetalization reaction is carried out in presence of acidic or basic materials. In order to compare the acidic activity of AlPO, VAPO and VAPSO materials, the acetalization of benzaldehyde was carried out over the above materials. The reaction can be written as shown below. C6H5CHO + 2CH3OH -------- C6H5CH(OCH3)2 + H2O (Benzaldehyde)
(Acetalide)
Conversion measured at different durations over AlPO4-5, VAPO-5 and VAPSO5 are presented in Fig. 5.2. Both AlPO4-5 and VAPO-5 appear to be of equal activity, while VAPSO-5 is much more active in keeping with its greater acidity. 5.6: MECHANISM OF TOLUENE OXIDATION Plausible routes for toluene oxidation to benzyl alcohol, benzaldehyde, o-cresol and p-cresol from VAPOs are given in Fig. 5.3 and Fig. 5.4. Both mechanism involves the fission of the C-H bond (in CH3) to form a benzylic radical which undergoes further transformation.
The formation of the benzylic radical initiated by peroxovanadium
species in the thermal case and is initiated by the applied voltage in electrocatalysis. Further reaction of benzylic radical with .OH either in solution or from the V-peroxo complexes leads to the side chain oxidation products. Isomerization of the benzylic radical and hydroxylation leads to the cresols.
132
Fig. 5.1: Reaction-schemes of (a) Toluene and (b) Cyclohexene oxidation with vanadium aluminophosphates.
133
Fig. 5.2: Time (hrs) vs conversion plot of benzaldehyde acetalization over a) AlPO4-5, b) VAPO-5 and c) VAPSO-5 catalysts.
134
Fig. 5.3: Proabable reaction mechanism (thermocatalysis) for toluene oxidation over VS1.
135
Fig. 5.4: Probable reaction mechanism (electrocatalysis) for toluene oxidation over VS-1. Table 5.1: Toluene oxidation over vanadium aluminophosphates.a Catalyst
Turn over Selectivityb number1
Benzyl
Benzal
Benzoic
alcohol
dehyde
acid
VAPO-5D
76
12
18
70
VAPSO-5A
60
9
36
55
VAPSO-5B
55
18
24
VAPO-31
106
24
6
a
o-cresol p-cresol
-
-
02
25
24
65
5
-
- Conditions: 0.1g catalyst, temperature 373 K, tertiary butyl hydroperoxide/toluene = 1
(molar ratio), duration = 22 h, chlorobenzene/toluene = 10 (molar ratio). a
turnover number = moles of toluene converted in 22h moles of V present in the catalyst
136
b
selectivity = weight percent of the specific product x 100 weight percent of all products
137
Table 5.2: A comparison of toluene oxidation in thermal and electrocatalysis.a Process
Electrocataly
Turnover
Conversionc
Selectivityd
numberb
(%)
Benzyl
Benzalde
alcohol
hyde
o-cresol
46
21
34
32
20
9
3
41
48
10
p-
Benzoic
cresol
acid
11
-
sis
Thermocatal ysis a
conditions:
Coulometric: Pellet = 0.6g (1:1, VAPO-5 and graphite), toluene/H2O2 or TBHP = 1 (molar ratio), toluene/acetonitrile = 10 (weight ratio), TBHP = tertiary butyl hydro peroxide, Current = 0.8mA, voltage = 1V, Duration = 12h, Coulombs of current passed = 201.5q, Temperature = room temperature.
Catalytic: 0.1g catalyst, toluene/H2O2 = 1 (molar ratio), toluene/ acetonitrile = 10 (molar ratio), temperature = 80oC, duration = 12h. b
turnover number = moles of toluene converted in 12h moles of V present in the catalyst
c
Conversion = weight percent of product x 100 weight percent of reactant and product
d
Selectivity
= weight percent of the specific product x 100 weight percent of all products
138
Table 5.3: Cyclohexene epoxidation over vanadium aluminophosphates.a Catalyst
b
Conversion (% TBHP)
VAPO-31 45 VAPO84 5C
Selectivity Cyclohexene cyclohexanone 32,3-epoxy epoxide cyclohexanol cyclochexanol 70 8 15 7 50 3 16 25
a
Conditions: catalyst = 0.1 g, temperature = 373 K, cyclohexene/tertiary butyl hydro peroxide = 1 (molar ratio); acetonitrile/cyclohexene = 10 (molar ratio).
139
Table 5.4: n-Hexane oxyfunctionalization over vanadium aluminophosphates.a Catalyst
Turnover
c
numberb
n-
n-
n-
n-
n-
n-
hexan
hexan
hexan
hexan
hexan
hexan
-1-ol
-2-ol
-3-ol
-1-al
-2-one -3-one
Selectivity (wt%)
VAPO-41
17
5
14
12
4
33
33
VAPO-31
54
3
11
7
9
34
36
a
Conditions: Catalyst = 0.5g, n-hexane = 5.6g, H2O2 = 2g, acetonitrile = 21.73g, temperature = 373K, time = 8h. b
c
Turnover number = moles of n-hexane converted in 8h moles of V present in the catalyst
Selectivity = weight percent of the specific product weight percent of total product
140
x 100
5.7: REFERENCES
1.
Prasad Rao, P.R.H., Ph.D. thesis entitled ―Synthesis, characterization and Catalytic properties of vanadium silicatemolecular sieves‖, submitted to Poona University, Pune, India, p.118 (1992).
2.
Thangaraj, A., Ph.D. Thesis entitled, ―Synthesis, characterization and catalytic properties of Titanium Silicate molecular sieves‖ submitted to Poona University, Pune, India, (1991).
3.
Sheldon, R.A., and Kochi, J.K., ―Metal Catalyzed Oxidation of Organic Compounds‖, Academic, New York (1981).
4.
Lyons, J.E., Appl. Ind. Catal., 3, 131 (1981).
5.
Meunier, B., Bull. Soc. Chim. Fr., 578 (1986).
6.
Herron, N., and Tolman, C.A. , J. Am. Chem. Soc., 109, 2837 (1987).
7.
Mimoun, H., Sarissiue, L., Daire, E., Postel, M., Fischer, J., and Weiss, R., J. Am. Chem. Soc., 105, 3101 (1983).
8.
Prasad Rao, P.R.H., and Ramaswamy, A.V., J. Chem. Soc., Chem. Commun., 1245 (1992).
9.
Prasad Rao, P.R.H., Ramaswamy, A.V. and Ratnasamy, P., J.Catal. 137, 225 (1992).
10.
Tatsumi, T., Nakamura, M., Negishi, S., and Tominaga, H., J. Chem. Soc. Chem. Commun., 476 (1990)
11.
Huy;brechts, D.R.C. Bruycker, L.D., and Jacobs, P.A. Nature, 345, 240 (1990).
141
CHAPTER VI ________________________________________________________________________ CONCLUSIONS
142
Hexamethyleneimine (HEM) has been used as a template for the synthesis of AlPO4-5, AlPO4-16, AlPO4-22 and SAPO-35 for the first time. During the studies on the synthesis of AlPO4-5, AlPO4-16, AlPO4-22 and SAPO-35, a novel crystalline layer material, AlPO4-L has been identified as an intermediate.
This is probably an
intermediate in all AlPO4 molecular sieve synthesis, when HEM is used as template. Physicochemical characterization shows that this is a lamellar type short range ordered layered aluminophosphate. Crystallization kinetics of AlPO4-5 using HEM template reveals that crystallization is faster compared to the conventional templates, viz., triethylamine and tripropylamine. The most crystalline sample was obtained at 473K in 4h from the reaction mixture of composition, Al2O3: P2O5: 1.16HEM: 45H2O. The addition of P in excess (1.2:1.0) leads to AlPO4-22, whereas excess of HEM (0.19M) gives AlPO4-16 structure. Changing the reaction medium to ethylene glycol and addition of silica (0.3SiO2) results in SAPO-35. Ethylene glycol is found to be superior to the aqueous medium in the synthesis of AlPO4-5 also. The location of Si in SAPO-35 was characterized by 29Si MASNMR and FT-IR of surface hydroxyl groups. Three types of bridging hydroxyl groups giving bands at 3580, 3604, 3631 cm-1 were observed and these have been assigned to different crystallographic positions. They have been assigned to locations at eight member ring cages near S6R, near D6R and inside D6R. New vanadium containing molecular sieves, VAPO-31, VAPO-41 and VAPSO-5 were synthesized. From the studies on vanadium incorporation in AlPO4-5, the following conclusions can be made. Tripropylamine is a more efficient template for vanadium incorporation compared to hexamethyleneimine. Vanadium incorporation is possible up to a critical value, above which dense phases are obtained. The crystalline VAPO-5 materials contain some amorphous vanadium oxides in the as-synthesized form, which increase on calcinations. Addition of silicon on the VAPO-5 synthesis gel enhances the crystallization rate and increases vanadium incorporation.
Vanadium replaces
phosphorous in the aluminophosphate framework. However in the presence of silicon, it has to compete to replace phosphorous.
Permanganometric titrations indicate the
presence of V5+ and V4+ ions in the as-synthesized samples.
Calcination at 823K
produces mostly V5+ ions. Elemental analysis show the sum of phosphorous and vanadium ions to be equal to aluminium.
The enhancement of the decomposition
temperature of the template in VAPO samples compared to AlPO suggests the presence of charged templates formed to compensate the negative charge produced when V4+ substitutes P5+ ions in the AlPO framework. Calcined VAPO-5 samples give a band at 950 cm-1 in the IR framework region showing the presence of (V=O) groups in the calcined samples.
27
Al NMR spectra reveal that most of the aluminium is in tetrahedral
co-ordination despite the presence of some unreacted alumina in the calcined VAPSO-5 samples.
29
Si shows most of the silicon to be in Si:4Si environment and a small amount
in Si:(2Al:2Si) and Si(3Al:Si) environments. UV-Vis spectra suggest that the vanadium species present in the as-synthesized samples are V5+(tetrahedral), V4+(tetrahedral), V4+(octahedral), while calcined samples contain V5+(tetrahedral) and V5+(square pyramidal) ions. Room temperature ESR spectra reveal eightline hyperfine splittings due to VO2+ species in a square pyramidal co-ordination. At liquid nitrogen temperature the intensity of the signal is enhanced four fold. This indicates that a large amount of VO2+ species is present in regular tetrahedral environment.
This signals disappears on
calcination and about 15% of the initial intensity is regained on reduction in H2. The amount reduced is equal to the amount of vanadium not leached out by ammonium acetate treatment and suggests that the framework vanadium undergoes redox transformation. Cyclic voltammetric studies using graphite paste modified electrodes suggest the presence of two distinct sets of redox couples in the cyclic voltammograms of calcined
143
samples of VAPO-5, VS-1, VS-2, V-EU-1, V-ZSM-22 and VAPSO-5. While a single redox couple is observed in the case of V-Al-, three redox couples are observed in VAPO-31. The two redox couples found in most samples are speculated to be V5+/V4+ in two different co-ordination environments, viz., a distorted tetrahedral symmetry and a square pyramidal symmetry , presumably present in different locations in the zeolite. The V5+ ions of square pyramidal symmetry are probably formed from the oxidation of V4+ ions present in the as-synthesized samples. VAPO-31 is found to be good catalyst for toluene oxidation, cyclochexene epoxidation and n-hexane oxifunctionalization.
Incorporation of silicon in VAPO-5
gives cresols as the major product while no cresols are formed on VAPO-5. VAPSO-5 samples are active in benzaldehyde acetalization. Electrocatalysis at room temperature leads to a seven-fold increase in conversion compared to the thermal reaction, while the selectivity remains unchanged.
144
LIST OF PUBLICATIONS 1.
Synthesis and characterization of a novel vanadium analogue of AlPO4-31. N. Venkatathri, S.G. Hegde, S. Sivasanker, J. Chem. Soc., Chem. Commun. (1995) 151.
2.
Cyclic voltammetric studies of vanadium in V-molecular sieves. N. Venkatathri, M.P. Vinod, K. Vijayamohanan and S. Sivasanker, J. Chem. Soc. Faraday Trans., 92 (1996) 473.
3.
Synthesis, characterization and catalytic properties of two novel vanado aluminosilicates with EU-1 and ZSM-22 structures. M. Chatterjee, D. Bhattacharya, N. Venkatathri and S. Sivasanker, Catal. Lett., 35 (1995) 313.
4.
Synthesis of SAPO-35 from non-aqueous gel. N. Venkatathri, S.G. Hegde, P.R. Rajamohanan and S. Sivasanker, J. Chem. Soc., Faraday Trans. (in press).
5.
Synthesis of AlPO4-5 using hexamethyleneimine template from aqueous and non-aqueous systems. N. Venkatathri, S.G. Hegde, V. Ramaswamy and S. Sivasanker, Zeolites (communicated).
6.
Synthesis and characterization of AlPO4-31 using long chain secondary amines. N. Venkatathri, N.E. Jacob, A.A. Behlekar and S.G. Hegde, Catalysis: Moder Trends, N.M. Gupta and D.K. Chakrabarty (Eds.), Narosha publishing house, New Delhi, India (1995) 72.
7.
Studies on the redox behaviour of vanadium containing molecular sieves by cyclic voltammetry. N. Venkatathri, M.P. Vinod, S. Sivasanker, and K. Vijayamohanan, Catalysis: Modern Trends, N.M. Gupta and D.K.Chakrabarty (Eds.), Narosha publishing house, New Delhi, India, (1995) 152.
8.
Thermal properties of ‗AFO‘ type aluminophosphate molecular sieves. N.Venkatathri, N.E. Jacob, and S.G. Hegde, 12th National Symposium on catalysis, Bhavnagar, India, December (1996).
9.
Studies on V-incorporation in AFI type materials. N. Venkatathri, S.G. Hegde, A. Behlekar and S. Sivasanker, Appl. Catal. (to be communicated).
10.
Synthesis, characterization and catalytic properties of VAPO-31 and VAPO-41 molecular sieves. N. Venkatathri, S.G. Hegde and S. Sivasanker, Ind. J. Chem. Section A (to be communicated).
11.
Isolation and characterization of a novel precursor phase for aluminophosphates. N. Venkatathri, S.G. Hegde, V. Ramaswamy, and S. Sivasanker, Zeolites ( to be communicated).
12.
Characterization of entrapped hexamethyleneimine in aluminophosphate molecular sieves. N. Venkatathri, S.G. Hegde, G. Sainker and S. Sivasanker, Microporous Materials (to be communicated).
13.
Studies on vanadium containing molecular sieves by cyclic voltammetry. Ind. J. Electro. Chem. (to be communicated).
14.
VS-1 graphite paste modified electrode catalysis on toluene oxidation. N. Venkatathri, M.P. Vinod, K. Vijayamohanan and S. Sivasanker, J. Chem. Soc., Chem. Commun. ( to be communicated).
15.
Identification of a novel aluminophosphate transformation, AlPO4-5 to AlPO4-22 through AlPO4-L. N. Venkatathri, V. Ramaswamy, S.G. Hegde, and S. Sivasanker, Microporous
145
Materials (to be communicated). 16.
V and Ti containing molecular sieves: Charging and discharging properties and effect on mass titration. N. Venkatathri, M.P. Vinod, K. Vijayamohanan and S. Sivasanker (manuscript in preparation).
146
147
